release(cli): Release new CLI version with support for api-version flag (#5640)

new CLI version with support for api-version flag

.github/ISSUE_TEMPLATE/privileged.yml

@@ -23,3 +23,7 @@ body:
     attributes:
       label: Issue Content
       description: Add the content of the issue here.
+  - type: markdown
+    attributes:
+      value: |
+        Community members should **NOT** work on Privileged issues unless these issues have been explicitly marked with a "help-wanted" tag.

.github/workflows/ci.yml

@@ -3,7 +3,7 @@ name: CI
 
 on:
   push:
-    branches: [main]
+    branches: [main, v1]
   pull_request:
 
 permissions:

.github/workflows/codespell.yml

@@ -34,7 +34,7 @@
             id: extract_ignore_words
     
           - name: Codespell
-            uses: codespell-project/actions-codespell@v2.0
+            uses: codespell-project/actions-codespell@v2.1
             with:
               skip: '*.ambr,*.lock,*.ipynb,*.yaml,*.zlib,*.css.map,*.js.map'
               ignore_words_list: ${{ steps.extract_ignore_words.outputs.ignore_words_list }}

docs/docs/agents/context.md

@@ -12,56 +12,64 @@ LangGraph provides **three** primary ways to supply context:
 
 | Type                                                                         | Description                                   | Mutable? | Lifetime                |
 |------------------------------------------------------------------------------|-----------------------------------------------|----------|-------------------------|
-| [**Config**](#config-static-context)                                         | data passed at the start of a run             | ❌        | per run                 |
+| [**Runtime Context**](#runtime-context)                                      | data passed at the start of a run             | ❌        | per run                 |
 | [**Short-term memory (State)**](#short-term-memory-mutable-context)          | dynamic data that can change during execution | ✅        | per run or conversation |
 | [**Long-term memory (Store)**](#long-term-memory-cross-conversation-context) | data that can be shared between conversations | ✅        | across conversations    |
 
-## Provide runtime context
+### Runtime Context
 
-### Config (static context)
+!!! note "`config['configurable']` -> `runtime.context`"
 
-Config is for immutable data like user metadata or API keys. Use
-when you have values that don't change mid-run.
+    In LangGraph < v1.0, static runtime context was passed via the `config['configurable']` key, paired with a `config_schema` argument
+    to `StateGraph` or `Pregel`. This is now deprecated and will be removed in v2.0.
 
-Specify configuration using a key called **"configurable"** which is reserved
-for this purpose:
+    As of LangGraph v1.0, the Runtime object is recommended to access static context and runtime-specific information like the store and stream writer.
+
+Runtime context is for immutable data like user metadata or API keys. Use this when you have values that don't change mid-run.
+
+Specify static context via the `context` argument to `invoke` / `stream`, which is reserved for this purpose:
 
 ```python
+@dataclass
+class ContextSchema:
+    user_name: str
+
 graph.invoke( # (1)!
     {"messages": [{"role": "user", "content": "hi!"}]}, # (2)!
     # highlight-next-line
-    config={"configurable": {"user_id": "user_123"}} # (3)!
+    context={"user_name": "John Smith"} # (3)!
 )
 ```
 
 1. This is the invocation of the agent or graph. The `invoke` method runs the underlying graph with the provided input.
 2. This example uses messages as an input, which is common, but your application may use different input structures.
-3. This is where you pass the configuration data. The `config` parameter allows you to provide additional context that the agent can use during its execution.
+3. This is where you pass the runtime data. The `context` parameter allows you to provide additional dependencies that the agent can use during its execution.
 
 === "Agent prompt"
 
     ```python
     from langchain_core.messages import AnyMessage
-    from langchain_core.runnables import RunnableConfig
+    from langgraph.runtime import get_runtime
     from langgraph.prebuilt.chat_agent_executor import AgentState
     from langgraph.prebuilt import create_react_agent
 
     # highlight-next-line
-    def prompt(state: AgentState, config: RunnableConfig) -> list[AnyMessage]:
-        user_name = config["configurable"].get("user_name")
-        system_msg = f"You are a helpful assistant. Address the user as {user_name}."
+    def prompt(state: AgentState) -> list[AnyMessage]:
+        runtime = get_runtime(ContextSchema)
+        system_msg = f"You are a helpful assistant. Address the user as {runtime.context.user_name}."
         return [{"role": "system", "content": system_msg}] + state["messages"]
 
     agent = create_react_agent(
         model="anthropic:claude-3-7-sonnet-latest",
         tools=[get_weather],
-        prompt=prompt
+        prompt=prompt,
+        context_schema=ContextSchema
     )
 
     agent.invoke(
         {"messages": [{"role": "user", "content": "what is the weather in sf"}]},
         # highlight-next-line
-        config={"configurable": {"user_name": "John Smith"}}
+        context={"user_name": "John Smith"}
     )
     ```
 
@@ -70,11 +78,11 @@ graph.invoke( # (1)!
 === "Workflow node"
 
     ```python
-    from langchain_core.runnables import RunnableConfig
+    from langgraph.runtime import Runtime
 
     # highlight-next-line
-    def node(state: State, config: RunnableConfig):
-        user_name = config["configurable"].get("user_name")
+    def node(state: State, config: Runtime[ContextSchema]):
+        user_name = runtime.context.user_name
         ...
     ```
 
@@ -83,14 +91,16 @@ graph.invoke( # (1)!
 === "In a tool"
 
     ```python
-    from langchain_core.runnables import RunnableConfig
+    from langgraph.runtime import get_runtime
 
     @tool
     # highlight-next-line
-    def get_user_info(config: RunnableConfig) -> str:
+    def get_user_email() -> str:
         """Retrieve user information based on user ID."""
-        user_id = config["configurable"].get("user_id")
-        return "User is John Smith" if user_id == "user_123" else "Unknown user"
+        # simulate fetching user info from a database
+        runtime = get_runtime(ContextSchema)
+        email = get_user_email_from_db(runtime.context.user_name)
+        return email
     ```
 
     See the [tool calling guide](../how-tos/tool-calling.md#configuration) for details.

docs/docs/cloud/deployment/egress.md

@@ -0,0 +1,119 @@
+# Egress for Subscription Metrics and Operational Metadata
+
+> **Important: Self Hosted Only** 
+> This section only applies to customers who are not running in offline mode and assumes you are using a self-hosted LangGraph Platform instance.
+> This does not apply to SaaS or Hybrid deployments.
+
+Self-Hosted LangGraph Platform instances store all information locally and will never send sensitive information outside of your network. We currently only track platform usage for billing purposes according to the entitlements in your order. In order to better remotely support our customers, we do require egress to `https://beacon.langchain.com`.
+
+In the future, we will be introducing support diagnostics to help us ensure that the LangGraph Platform is running at an optimal level within your environment.
+
+> **Warning**  
+> **This will require egress to `https://beacon.langchain.com` from your network.**
+> **If using an API key, you will also need to allow egress to `https://api.smith.langchain.com` or `https://eu.api.smith.langchain.com` for API key verification.**
+
+Generally, data that we send to Beacon can be categorized as follows:
+
+- **Subscription Metrics**
+  - Subscription metrics are used to determine level of access and utilization of LangSmith. This includes, but are not limited to:
+    - Nodes Executed
+    - Runs Executed
+    - License Key Verification
+- **Operational Metadata**
+  - This metadata will contain and collect the above subscription metrics to assist with remote support, allowing the LangChain team to diagnose and troubleshoot performance issues more effectively and proactively.
+
+## Example Payloads
+
+In an effort to maximize transparency, we provide sample payloads here:
+
+### License Verification (If using an Enterprise License)
+
+**Endpoint:**
+
+`POST beacon.langchain.com/v1/beacon/verify`
+
+**Request:**
+
+```json
+{
+  "license": "<YOUR_LICENSE_KEY>"
+}
+```
+
+**Response:**
+
+```json
+{
+  "token": "Valid JWT" // Short-lived JWT token to avoid repeated license checks
+}
+```
+
+### Api Key Verification (If using a LangSmith API Key)
+
+**Endpoint:**
+`POST api.smith.langchain.com/auth`
+
+**Request:**
+
+```json
+"Headers": {
+  X-Api-Key: <YOUR_API_KEY>
+}
+```
+
+**Response:**
+
+```json
+{
+  "org_config": {
+    "org_id": "3a1c2b6f-4430-4b92-8a5b-79b8b567bbc1",
+    ... // Additional organization details
+  }
+}
+```
+
+### Usage Reporting
+
+**Endpoint:**
+
+`POST beacon.langchain.com/v1/metadata/submit`
+
+**Request:**
+
+```json
+{
+  "license": "<YOUR_LICENSE_KEY>",
+  "from_timestamp": "2025-01-06T09:00:00Z",
+  "to_timestamp": "2025-01-06T10:00:00Z",
+  "tags": {
+    "langgraph.python.version": "0.1.0",
+    "langgraph_api.version": "0.2.0",
+    "langgraph.platform.revision": "abc123",
+    "langgraph.platform.variant": "standard",
+    "langgraph.platform.host": "host-1",
+    "langgraph.platform.tenant_id": "3a1c2b6f-4430-4b92-8a5b-79b8b567bbc1",
+    "langgraph.platform.project_id": "c5b5f53a-4716-4326-8967-d4f7f7799735",
+    "langgraph.platform.plan": "enterprise",
+    "user_app.uses_indexing": "true",
+    "user_app.uses_custom_app": "false",
+    "user_app.uses_custom_auth": "true",
+    "user_app.uses_thread_ttl": "true",
+    "user_app.uses_store_ttl": "false"
+  },
+  "measures": {
+    "langgraph.platform.runs": 150,
+    "langgraph.platform.nodes": 450
+  },
+  "logs": []
+}
+```
+
+**Response:**
+
+```json
+"204 No Content"
+```
+
+## Our Commitment
+
+LangChain will not store any sensitive information in the Subscription Metrics or Operational Metadata. Any data collected will not be shared with a third party. If you have any concerns about the data being sent, please reach out to your account team.

docs/docs/cloud/deployment/self_hosted_control_plane.md

@@ -23,6 +23,8 @@ Before deploying, review the [conceptual guide for the Self-Hosted Control Plane
 
         kubectl get storageclass
 
+1. Egress to `https://beacon.langchain.com` from your network. This is required for license verification and usage reporting if not running in air-gapped mode. See the [Egress documentation](../../cloud/deployment/egress.md) for more details.
+
 ## Setup
 
 1. As part of configuring your Self-Hosted LangSmith instance, you enable the `langgraphPlatform` option. This will provision a few key resources.

docs/docs/cloud/deployment/setup.md

@@ -108,11 +108,11 @@ from langgraph.graph import StateGraph, END, START
 from my_agent.utils.nodes import call_model, should_continue, tool_node # import nodes
 from my_agent.utils.state import AgentState # import state
 
-# Define the config
-class GraphConfig(TypedDict):
+# Define the runtime context
+class GraphContext(TypedDict):
     model_name: Literal["anthropic", "openai"]
 
-workflow = StateGraph(AgentState, config_schema=GraphConfig)
+workflow = StateGraph(AgentState, context_schema=GraphContext)
 workflow.add_node("agent", call_model)
 workflow.add_node("action", tool_node)
 workflow.add_edge(START, "agent")

docs/docs/cloud/deployment/setup_pyproject.md

@@ -121,11 +121,11 @@ from langgraph.graph import StateGraph, END, START
 from my_agent.utils.nodes import call_model, should_continue, tool_node # import nodes
 from my_agent.utils.state import AgentState # import state
 
-# Define the config
-class GraphConfig(TypedDict):
+# Define the runtime context
+class GraphContext(TypedDict):
     model_name: Literal["anthropic", "openai"]
 
-workflow = StateGraph(AgentState, config_schema=GraphConfig)
+workflow = StateGraph(AgentState, context_schema=GraphContext)
 workflow.add_node("agent", call_model)
 workflow.add_node("action", tool_node)
 workflow.add_edge(START, "agent")

docs/docs/cloud/deployment/standalone_container.md

@@ -24,6 +24,7 @@ Before deploying, review the [conceptual guide for the Standalone Container](../
     1. `LANGSMITH_API_KEY`: (if using [Lite](../../concepts/langgraph_server.md#server-versions)) LangSmith API key. This will be used to authenticate ONCE at server start up.
     1. `LANGGRAPH_CLOUD_LICENSE_KEY`: (if using [Enterprise](../../concepts/langgraph_data_plane.md#licensing)) LangGraph Platform license key. This will be used to authenticate ONCE at server start up.
     1. `LANGSMITH_ENDPOINT`: To send traces to a [self-hosted LangSmith](https://docs.smith.langchain.com/self_hosting) instance, set `LANGSMITH_ENDPOINT` to the hostname of the self-hosted LangSmith instance.
+1. Egress to `https://beacon.langchain.com` from your network. This is required for license verification and usage reporting if not running in air-gapped mode. See the [Egress documentation](../../cloud/deployment/egress.md) for more details.
 
 ## Kubernetes (Helm)
 

docs/docs/cloud/how-tos/add-human-in-the-loop.md

@@ -30,9 +30,7 @@ To review, edit, and approve tool calls in an agent or workflow, use LangGraph's
     # > [
     # >     {
     # >         'value': {'text_to_revise': 'original text'},
-    # >         'resumable': True,
-    # >         'ns': ['human_node:fc722478-2f21-0578-c572-d9fc4dd07c3b'],
-    # >         'when': 'during'
+    # >         'id': '...',
     # >     }
     # > ]
 
@@ -203,9 +201,7 @@ To review, edit, and approve tool calls in an agent or workflow, use LangGraph's
         # > [
         # >     {
         # >         'value': {'text_to_revise': 'original text'},
-        # >         'resumable': True,
-        # >         'ns': ['human_node:fc722478-2f21-0578-c572-d9fc4dd07c3b'],
-        # >         'when': 'during'
+        # >         'id': '...',
         # >     }
         # > ]
 

docs/docs/cloud/how-tos/configuration_cloud.md

@@ -2,21 +2,20 @@
 
 In this guide we will show how to create, configure, and manage an [assistant](../../concepts/assistants.md).
 
-First, as a brief refresher on the concept of configurations, consider the following simple `call_model` node and configuration schema. Observe that this node tries to read and use the `model_name` as defined by the `config` object's `configurable`.
+First, as a brief refresher on the concept of runtime context, consider the following simple `call_model` node and context schema. Observe that this node tries to read and use the `model_provider` as defined by the `Runtime` object's `context` property.
 
 === "Python"
 
     ```python
+    @dataclass
+    class ContextSchema:
+        llm_provider: str = "anthropic"
 
-    class ConfigSchema(TypedDict):
-        model_name: str
+    builder = StateGraph(AgentState, context_schema=ContextSchema)
 
-    builder = StateGraph(AgentState, config_schema=ConfigSchema)
-
-    def call_model(state, config):
+    def call_model(state, runtime: Runtime[ContextSchema]):
         messages = state["messages"]
-        model_name = config.get('configurable', {}).get("model_name", "anthropic")
-        model = _get_model(model_name)
+        model = _get_model(runtime.context.llm_provider)
         response = model.invoke(messages)
         # We return a list, because this will get added to the existing list
         return {"messages": [response]}
@@ -44,7 +43,7 @@ First, as a brief refresher on the concept of configurations, consider the follo
     }
     ```
 
-For more information on configurations, [see here](../../concepts/low_level.md#configuration).
+For more information on runtime context, [see here](../../concepts/low_level.md#runtime-context).
 
 ## Create an assistant
 

docs/docs/cloud/how-tos/generative_ui_react.md

@@ -30,17 +30,33 @@ export default {
 
 Next, define your UI components in your `langgraph.json` configuration:
 
-```json
-{
-  "node_version": "20",
-  "graphs": {
-    "agent": "./src/agent/index.ts:graph"
-  },
-  "ui": {
-    "agent": "./src/agent/ui.tsx"
-  }
-}
-```
+=== "Python agent"
+
+    ```json title="langgraph.json"
+    {
+      "node_version": "20",
+      "graphs": {
+        "agent": "./src/agent.py:graph"
+      },
+      "ui": {
+        "agent": "./src/agent/ui.tsx"
+      }
+    }
+    ```
+
+=== "JS agent"
+
+    ```json title="langgraph.json"
+    {
+      "node_version": "20",
+      "graphs": {
+        "agent": "./src/agent/index.ts:graph"
+      },
+      "ui": {
+        "agent": "./src/agent/ui.tsx"
+      }
+    }
+    ```
 
 The `ui` section points to the UI components that will be used by graphs. By default, we recommend using the same key as the graph name, but you can split out the components however you like, see [Customise the namespace of UI components](#customise-the-namespace-of-ui-components) for more details.
 

docs/docs/cloud/how-tos/webhooks.md

@@ -140,6 +140,22 @@ https://my-server.app/my-webhook-endpoint?token=YOUR_SECRET_TOKEN
 
 Your server should extract and validate this token before processing requests.
 
+## Disable webhooks
+
+As of `langgraph-api>=0.2.78`, developers can disable webhooks in the `langgraph.json` file:
+
+```json
+{
+  "http": {
+    "disable_webhooks": true
+  }
+}
+```
+
+This feature is primarily intended for self-hosted deployments, where platform administrators or developers may prefer to disable webhooks to simplify their security posture—especially if they are not configuring firewall rules or other network controls. Disabling webhooks helps prevent untrusted payloads from being sent to internal endpoints.
+
+For full configuration details, refer to the [configuration file reference](https://langchain-ai.github.io/langgraph/cloud/reference/cli/?h=disable_webhooks#configuration-file).
+
 ## Test webhooks
 
 You can test your webhook using online services like:

docs/docs/cloud/reference/langgraph_server_changelog.md

@@ -4,6 +4,18 @@
 
 ---
 
+## v0.2.98 (2025-07-19)
+- Added langgraph node context for improved log filtering and trace visibility.
+
+## v0.2.97 (2025-07-19)
+- Fixed scheduling issue with ckpt ingestion worker that occurred on isolated background loops.
+- Ensured queue worker starts only after all migrations have completed.
+- Added more detailed error messages for thread state issues and improved response handling when state updates fail.
+- Exposed interrupt ID while retrieving thread state for enhanced API response details.
+
+## v0.2.96 (2025-07-17)
+- Added a fallback mechanism for configurable header patterns to handle exclude/include settings more effectively.
+
 ## v0.2.95 (2025-07-17)
 - Avoided setting the future if it is already done to prevent redundant operations.
 - Resolved compatibility errors in CI by switching from `typing.TypedDict` to `typing_extensions.TypedDict` for Python versions below 3.12.

docs/docs/concepts/assistants.md

@@ -1,6 +1,6 @@
 # Assistants
 
-**Assistants** allow you to manage configurations (like prompts, LLM selection, tools) separately from your graph's core logic, enabling rapid changes that don't alter the graph architecture. It is a way to create multiple specialized versions of the same graph architecture, each optimized for different use cases through configuration variations rather than structural changes.
+**Assistants** allow you to manage configurations (like prompts, LLM selection, tools) separately from your graph's core logic, enabling rapid changes that don't alter the graph architecture. It is a way to create multiple specialized versions of the same graph architecture, each optimized for different use cases through context/configuration variations rather than structural changes.
 
 For example, imagine a general-purpose writing agent built on a common graph architecture. While the structure remains the same, different writing styles—such as blog posts and tweets—require tailored configurations to optimize performance. To support these variations, you can create multiple assistants (e.g., one for blogs and another for tweets) that share the underlying graph but differ in model selection and system prompt.
 
@@ -14,8 +14,8 @@ The LangGraph Cloud API provides several endpoints for creating and managing ass
 
 ## Configuration
 
-Assistants build on the LangGraph open source concept of [configuration](low_level.md#configuration).
-While configuration is available in the open source LangGraph library, assistants are only present in [LangGraph Platform](langgraph_platform.md). This is due to the fact that assistants are tightly coupled to your deployed graph. Upon deployment, LangGraph Server will automatically create a default assistant for each graph using the graph's default configuration settings.
+Assistants build on the LangGraph open source concepts of configuration and [runtime context](low_level.md#runtime-context).
+While these features are available in the open source LangGraph library, assistants are only present in [LangGraph Platform](langgraph_platform.md). This is due to the fact that assistants are tightly coupled to your deployed graph. Upon deployment, LangGraph Server will automatically create a default assistant for each graph using the graph's default context and configuration settings.
 
 In practice, an assistant is just an _instance_ of a graph with a specific configuration. Therefore, multiple assistants can reference the same graph but can contain different configurations (e.g. prompts, models, tools). The LangGraph Server API provides several endpoints for creating and managing assistants. See the [API reference](../cloud/reference/api/api_ref.html) and [this how-to](../cloud/how-tos/configuration_cloud.md) for more details on how to create assistants.
 
@@ -26,6 +26,6 @@ Once you've created an assistant, subsequent edits to that assistant will create
 
 ## Execution
 
-A **run** is an invocation of an assistant. Each run may have its own input, configuration, and metadata, which may affect execution and output of the underlying graph. A run can optionally be executed on a [thread](./persistence.md#threads).
+A **run** is an invocation of an assistant. Each run may have its own input, configuration, context, and metadata, which may affect execution and output of the underlying graph. A run can optionally be executed on a [thread](./persistence.md#threads).
 
 The LangGraph Platform API provides several endpoints for creating and managing runs. See the [API reference](../cloud/reference/api/api_ref.html#tag/thread-runs/) for more details.

docs/docs/concepts/deployment_options.md

@@ -10,7 +10,7 @@ search:
 There are two free options for deploying LangGraph applications via the LangGraph Server:
 
 1. [Local](../tutorials/langgraph-platform/local-server.md): Deploy for local testing and development. 
-1. [Standalone Container (Lite)](../concepts/langgraph_standalone_container.md): A limited version of Standalone Container for deployments unlikely to see more that 1 million node executions per year and that do not need crons and other enterprise features. Standalone Container (Lite) deployment option is free with a LangSmith API key.
+1. [Standalone Container (Lite)](../concepts/langgraph_standalone_container.md): A limited version of Standalone Container for deployments unlikely to see more than 1 million node executions per year and that do not need crons and other enterprise features. Standalone Container (Lite) deployment option is free with a LangSmith API key.
 
 ## Production deployment
 

docs/docs/concepts/durable_execution.md

@@ -48,7 +48,7 @@ If a [node](./low_level.md#nodes) contains multiple operations, you may find it
     from typing_extensions import TypedDict
     import uuid
 
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
     from langgraph.graph import StateGraph, START, END
     import requests
 
@@ -74,7 +74,7 @@ If a [node](./low_level.md#nodes) contains multiple operations, you may find it
     builder.add_edge("call_api", END)
 
     # Specify a checkpointer
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
 
     # Compile the graph with the checkpointer
     graph = builder.compile(checkpointer=checkpointer)
@@ -94,7 +94,7 @@ If a [node](./low_level.md#nodes) contains multiple operations, you may find it
     from typing_extensions import TypedDict
     import uuid
 
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
     from langgraph.func import task
     from langgraph.graph import StateGraph, START, END
     import requests
@@ -129,7 +129,7 @@ If a [node](./low_level.md#nodes) contains multiple operations, you may find it
     builder.add_edge("call_api", END)
 
     # Specify a checkpointer
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
 
     # Compile the graph with the checkpointer
     graph = builder.compile(checkpointer=checkpointer)

docs/docs/concepts/functional_api.md

@@ -39,7 +39,7 @@ Here are some key differences:
 Below we demonstrate a simple application that writes an essay and [interrupts](human_in_the_loop.md) to request human review.
 
 ```python
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.func import entrypoint, task
 from langgraph.types import interrupt
 
@@ -50,7 +50,7 @@ def write_essay(topic: str) -> str:
     time.sleep(1) # A placeholder for a long-running task.
     return f"An essay about topic: {topic}"
 
-@entrypoint(checkpointer=MemorySaver())
+@entrypoint(checkpointer=InMemorySaver())
 def workflow(topic: str) -> dict:
     """A simple workflow that writes an essay and asks for a review."""
     essay = write_essay("cat").result()
@@ -79,51 +79,54 @@ def workflow(topic: str) -> dict:
     ```python
     import time
     import uuid
-
     from langgraph.func import entrypoint, task
     from langgraph.types import interrupt
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
+
 
     @task
     def write_essay(topic: str) -> str:
         """Write an essay about the given topic."""
-        time.sleep(1) # This is a placeholder for a long-running task.
+        time.sleep(1)  # This is a placeholder for a long-running task.
         return f"An essay about topic: {topic}"
 
-    @entrypoint(checkpointer=MemorySaver())
+    @entrypoint(checkpointer=InMemorySaver())
     def workflow(topic: str) -> dict:
         """A simple workflow that writes an essay and asks for a review."""
         essay = write_essay("cat").result()
-        is_approved = interrupt({
-            # Any json-serializable payload provided to interrupt as argument.
-            # It will be surfaced on the client side as an Interrupt when streaming data
-            # from the workflow.
-            "essay": essay, # The essay we want reviewed.
-            # We can add any additional information that we need.
-            # For example, introduce a key called "action" with some instructions.
-            "action": "Please approve/reject the essay",
-        })
-
+        is_approved = interrupt(
+            {
+                # Any json-serializable payload provided to interrupt as argument.
+                # It will be surfaced on the client side as an Interrupt when streaming data
+                # from the workflow.
+                "essay": essay,  # The essay we want reviewed.
+                # We can add any additional information that we need.
+                # For example, introduce a key called "action" with some instructions.
+                "action": "Please approve/reject the essay",
+            }
+        )
         return {
-            "essay": essay, # The essay that was generated
-            "is_approved": is_approved, # Response from HIL
+            "essay": essay,  # The essay that was generated
+            "is_approved": is_approved,  # Response from HIL
         }
 
+
     thread_id = str(uuid.uuid4())
-
-    config = {
-        "configurable": {
-            "thread_id": thread_id
-        }
-    }
-
+    config = {"configurable": {"thread_id": thread_id}}
     for item in workflow.stream("cat", config):
         print(item)
-    ```
-
-    ```pycon
-    {'write_essay': 'An essay about topic: cat'}
-    {'__interrupt__': (Interrupt(value={'essay': 'An essay about topic: cat', 'action': 'Please approve/reject the essay'}, resumable=True, ns=['workflow:f7b8508b-21c0-8b4c-5958-4e8de74d2684'], when='during'),)}
+    # > {'write_essay': 'An essay about topic: cat'}
+    # > {
+    # >     '__interrupt__': (
+    # >        Interrupt(
+    # >            value={
+    # >                'essay': 'An essay about topic: cat',
+    # >                'action': 'Please approve/reject the essay'
+    # >            },
+    # >            id='b9b2b9d788f482663ced6dc755c9e981'
+    # >        ),
+    # >    )
+    # > }
     ```
 
     An essay has been written and is ready for review. Once the review is provided, we can resume the workflow:

docs/docs/concepts/langgraph_control_plane.md

@@ -119,6 +119,11 @@ These metrics are displayed as charts in the Control Plane UI.
 
 ### LangSmith Integration
 
-A [LangSmith](https://docs.smith.langchain.com/) tracing project is automatically created for each deployment. The tracing project has the same name as the deployment. When creating a deployment, the `LANGCHAIN_TRACING` and `LANGSMITH_API_KEY`/`LANGCHAIN_API_KEY` environment variables do not need to be specified; they are set automatically by the control plane.
+A [LangSmith](https://docs.smith.langchain.com/) tracing project and LangSmith API key are automatically created for each deployment. The deployment uses the API key to automatically send traces to LangSmith.
 
-When a deployment is deleted, the traces and the tracing project are not deleted.
+- The tracing project has the same name as the deployment.
+- The API key has the description `LangGraph Platform: <deployment_name>`.
+- The API key is never revealed and cannot be deleted manually.
+- When creating a deployment, the `LANGCHAIN_TRACING` and `LANGSMITH_API_KEY`/`LANGCHAIN_API_KEY` environment variables do not need to be specified; they are set automatically by the control plane.
+
+When a deployment is deleted, the traces and the tracing project are not deleted. However, the API will be deleted when the deployment is deleted.

docs/docs/concepts/low_level.md

@@ -192,35 +192,48 @@ class State(MessagesState):
 
 ## Nodes
 
-In LangGraph, nodes are typically python functions (sync or async) where the **first** positional argument is the [state](#state), and (optionally), the **second** positional argument is a "config", containing optional [configurable parameters](#configuration) (such as a `thread_id`).
+In LangGraph, nodes are Python functions (either synchronous or asynchronous) that accept the following arguments:
+
+1. `state`: The [state](#state) of the graph
+2. `config`: A `RunnableConfig` object that contains configuration information like `thread_id` and tracing information like `tags`
+3. `runtime`: A `Runtime` object that contains [runtime `context`](#runtime-context) and other information like `store` and `stream_writer`
+    
 
 Similar to `NetworkX`, you add these nodes to a graph using the [add_node][langgraph.graph.StateGraph.add_node] method:
 
 ```python
+from dataclasses import dataclass
 from typing_extensions import TypedDict
 
 from langchain_core.runnables import RunnableConfig
 from langgraph.graph import StateGraph
+from langgraph.runtime import Runtime
 
 class State(TypedDict):
     input: str
     results: str
 
+@dataclass
+class Context:
+    user_id: str
+
 builder = StateGraph(State)
 
+def plain_node(state: State):
+    return state
 
-def my_node(state: State, config: RunnableConfig):
-    print("In node: ", config["configurable"]["user_id"])
+def node_with_runtime(state: State, runtime: Runtime[Context]):
+    print("In node: ", runtime.context.user_id)
+    return {"results": f"Hello, {state['input']}!"}
+
+def node_with_config(state: State, config: RunnableConfig):
+    print("In node with thread_id: ", config["configurable"]["thread_id"])
     return {"results": f"Hello, {state['input']}!"}
 
 
-# The second argument is optional
-def my_other_node(state: State):
-    return state
-
-
-builder.add_node("my_node", my_node)
-builder.add_node("other_node", my_other_node)
+builder.add_node("plain_node", plain_node)
+builder.add_node("node_with_runtime", node_with_runtime)
+builder.add_node("node_with_config", node_with_config)
 ...
 ```
 
@@ -459,33 +472,32 @@ LangGraph can easily handle migrations of graph definitions (nodes, edges, and s
 - State keys that are renamed lose their saved state in existing threads
 - State keys whose types change in incompatible ways could currently cause issues in threads with state from before the change -- if this is a blocker please reach out and we can prioritize a solution.
 
-## Configuration
+## Runtime Context
 
-When creating a graph, you can also mark that certain parts of the graph are configurable. This is commonly done to enable easily switching between models or system prompts. This allows you to create a single "cognitive architecture" (the graph) but have multiple different instance of it.
-
-You can optionally specify a `config_schema` when creating a graph.
+When creating a graph, you can specify a `context_schema` for runtime context passed to nodes. This is useful for passing
+information to nodes that is not part of the graph state. For example, you might want to pass dependencies such as model name or a database connection.
 
 ```python
-class ConfigSchema(TypedDict):
-    llm: str
+@dataclass
+class ContextSchema:
+    llm_provider: str = "openai"
 
-graph = StateGraph(State, config_schema=ConfigSchema)
+graph = StateGraph(State, context_schema=ContextSchema)
 ```
 
-You can then pass this configuration into the graph using the `configurable` config field.
+You can then pass this context into the graph using the `context` parameter of the `invoke` method.
 
 ```python
-config = {"configurable": {"llm": "anthropic"}}
-
-graph.invoke(inputs, config=config)
+graph.invoke(inputs, context={"llm_provider": "anthropic"})
 ```
 
-You can then access and use this configuration inside a node or conditional edge:
+You can then access and use this context inside a node or conditional edge:
 
 ```python
-def node_a(state, config):
-    llm_type = config.get("configurable", {}).get("llm", "openai")
-    llm = get_llm(llm_type)
+from langgraph.runtime import Runtime
+
+def node_a(state: State, runtime: Runtime[ContextSchema]):
+    llm = get_llm(runtime.context.llm_provider)
     ...
 ```
 
@@ -496,7 +508,7 @@ See [this guide](../how-tos/graph-api.md#add-runtime-configuration) for a full b
 The recursion limit sets the maximum number of [super-steps](#graphs) the graph can execute during a single execution. Once the limit is reached, LangGraph will raise `GraphRecursionError`. By default this value is set to 25 steps. The recursion limit can be set on any graph at runtime, and is passed to `.invoke`/`.stream` via the config dictionary. Importantly, `recursion_limit` is a standalone `config` key and should not be passed inside the `configurable` key as all other user-defined configuration. See the example below:
 
 ```python
-graph.invoke(inputs, config={"recursion_limit": 5, "configurable":{"llm": "anthropic"}})
+graph.invoke(inputs, config={"recursion_limit": 5}, context={"llm": "anthropic"})
 ```
 
 Read [this how-to](https://langchain-ai.github.io/langgraph/how-tos/recursion-limit/) to learn more about how the recursion limit works.

docs/docs/concepts/persistence.md

@@ -487,12 +487,12 @@ If you want to fallback to pickle for objects not currently supported by our msg
 you can use the `pickle_fallback` argument of the `JsonPlusSerializer`:
 
 ```python
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.checkpoint.serde.jsonplus import JsonPlusSerializer
 
 # ... Define the graph ...
 graph.compile(
-    checkpointer=MemorySaver(serde=JsonPlusSerializer(pickle_fallback=True))
+    checkpointer=InMemorySaver(serde=JsonPlusSerializer(pickle_fallback=True))
 )
 ```
 

docs/docs/how-tos/autogen-integration-functional.ipynb

@@ -77,7 +77,7 @@
    "metadata": {},
    "outputs": [
     {
-     "name": "stdin",
+     "name": "stdout",
      "output_type": "stream",
      "text": [
       "OPENAI_API_KEY:  ········\n"
@@ -165,7 +165,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": null,
    "id": "d129e4e1-3766-429a-b806-cde3d8bc0469",
    "metadata": {},
    "outputs": [],
@@ -173,7 +173,7 @@
     "from langchain_core.messages import convert_to_openai_messages, BaseMessage\n",
     "from langgraph.func import entrypoint, task\n",
     "from langgraph.graph import add_messages\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "\n",
     "\n",
     "@task\n",
@@ -192,7 +192,7 @@
     "\n",
     "\n",
     "# add short-term memory for storing conversation history\n",
-    "checkpointer = MemorySaver()\n",
+    "checkpointer = InMemorySaver()\n",
     "\n",
     "\n",
     "@entrypoint(checkpointer=checkpointer)\n",
@@ -222,12 +222,12 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001B[33muser_proxy\u001B[0m (to assistant):\n",
+      "\u001b[33muser_proxy\u001b[0m (to assistant):\n",
       "\n",
       "Find numbers between 10 and 30 in fibonacci sequence\n",
       "\n",
       "--------------------------------------------------------------------------------\n",
-      "\u001B[33massistant\u001B[0m (to user_proxy):\n",
+      "\u001b[33massistant\u001b[0m (to user_proxy):\n",
       "\n",
       "To find numbers between 10 and 30 in the Fibonacci sequence, we can generate the Fibonacci sequence and check which numbers fall within this range. Here's a plan:\n",
       "\n",
@@ -253,9 +253,9 @@
       "This script will print the Fibonacci numbers between 10 and 30. Please execute the code to see the result.\n",
       "\n",
       "--------------------------------------------------------------------------------\n",
-      "\u001B[31m\n",
-      ">>>>>>>> EXECUTING CODE BLOCK 0 (inferred language is python)...\u001B[0m\n",
-      "\u001B[33muser_proxy\u001B[0m (to assistant):\n",
+      "\u001b[31m\n",
+      ">>>>>>>> EXECUTING CODE BLOCK 0 (inferred language is python)...\u001b[0m\n",
+      "\u001b[33muser_proxy\u001b[0m (to assistant):\n",
       "\n",
       "exitcode: 0 (execution succeeded)\n",
       "Code output: \n",
@@ -264,7 +264,7 @@
       "\n",
       "\n",
       "--------------------------------------------------------------------------------\n",
-      "\u001B[33massistant\u001B[0m (to user_proxy):\n",
+      "\u001b[33massistant\u001b[0m (to user_proxy):\n",
       "\n",
       "The Fibonacci numbers between 10 and 30 are 13 and 21. \n",
       "\n",
@@ -318,7 +318,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001B[33muser_proxy\u001B[0m (to assistant):\n",
+      "\u001b[33muser_proxy\u001b[0m (to assistant):\n",
       "\n",
       "Multiply the last number by 3\n",
       "Context: \n",
@@ -334,7 +334,7 @@
       "TERMINATE\n",
       "\n",
       "--------------------------------------------------------------------------------\n",
-      "\u001B[33massistant\u001B[0m (to user_proxy):\n",
+      "\u001b[33massistant\u001b[0m (to user_proxy):\n",
       "\n",
       "The last number in the Fibonacci sequence between 10 and 30 is 21. Multiplying 21 by 3 gives:\n",
       "\n",

docs/docs/how-tos/autogen-integration.md

@@ -75,7 +75,7 @@ We will now create a LangGraph chatbot graph that calls AutoGen agent.
 ```python
 from langchain_core.messages import convert_to_openai_messages
 from langgraph.graph import StateGraph, MessagesState, START
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 
 def call_autogen_agent(state: MessagesState):
     # Convert LangGraph messages to OpenAI format for AutoGen
@@ -101,7 +101,7 @@ def call_autogen_agent(state: MessagesState):
     return {"messages": {"role": "assistant", "content": final_content}}
 
 # Create the graph with memory for persistence
-checkpointer = MemorySaver()
+checkpointer = InMemorySaver()
 
 # Build the graph
 builder = StateGraph(MessagesState)
@@ -228,7 +228,7 @@ my-autogen-agent/
     import autogen
     from langchain_core.messages import convert_to_openai_messages
     from langgraph.graph import StateGraph, MessagesState, START
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
     # AutoGen configuration
     config_list = [{"model": "gpt-4o", "api_key": os.environ["OPENAI_API_KEY"]}]
@@ -276,7 +276,7 @@ my-autogen-agent/
 
     # Create and compile the graph
     def create_graph():
-        checkpointer = MemorySaver()
+        checkpointer = InMemorySaver()
         builder = StateGraph(MessagesState)
         builder.add_node("autogen", call_autogen_agent)
         builder.add_edge(START, "autogen")
@@ -290,7 +290,7 @@ my-autogen-agent/
 
     ```
     langgraph>=0.1.0
-    pyautogen>=0.2.0
+    ag2>=0.2.0
     langchain-core>=0.1.0
     langchain-openai>=0.0.5
     ```

docs/docs/how-tos/cross-thread-persistence-functional.ipynb

@@ -167,7 +167,7 @@
     "from langchain_core.messages import BaseMessage\n",
     "from langgraph.func import entrypoint, task\n",
     "from langgraph.graph import add_messages\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from langgraph.store.base import BaseStore\n",
     "\n",
     "\n",
@@ -192,7 +192,7 @@
     "\n",
     "\n",
     "# NOTE: we're passing the store object here when creating a workflow via entrypoint()\n",
-    "@entrypoint(checkpointer=MemorySaver(), store=in_memory_store)\n",
+    "@entrypoint(checkpointer=InMemorySaver(), store=in_memory_store)\n",
     "def workflow(\n",
     "    inputs: list[BaseMessage],\n",
     "    *,\n",

docs/docs/how-tos/graph-api.md

@@ -514,12 +514,12 @@ To add runtime configuration:
 See below for a simple example:
 
 ```python
-from langchain_core.runnables import RunnableConfig
 from langgraph.graph import END, StateGraph, START
+from langgraph.runtime import Runtime
 from typing_extensions import TypedDict
 
 # 1. Specify config schema
-class ConfigSchema(TypedDict):
+class ContextSchema(TypedDict):
     my_runtime_value: str
 
 # 2. Define a graph that accesses the config in a node
@@ -527,18 +527,18 @@ class State(TypedDict):
     my_state_value: str
 
 # highlight-next-line
-def node(state: State, config: RunnableConfig):
+def node(state: State, runtime: Runtime[ContextSchema]):
     # highlight-next-line
-    if config["configurable"]["my_runtime_value"] == "a":
+    if runtime.context["my_runtime_value"] == "a":
         return {"my_state_value": 1}
         # highlight-next-line
-    elif config["configurable"]["my_runtime_value"] == "b":
+    elif runtime.context["my_runtime_value"] == "b":
         return {"my_state_value": 2}
     else:
         raise ValueError("Unknown values.")
 
 # highlight-next-line
-builder = StateGraph(State, config_schema=ConfigSchema)
+builder = StateGraph(State, context_schema=ContextSchema)
 builder.add_node(node)
 builder.add_edge(START, "node")
 builder.add_edge("node", END)
@@ -547,9 +547,9 @@ graph = builder.compile()
 
 # 3. Pass in configuration at runtime:
 # highlight-next-line
-print(graph.invoke({}, {"configurable": {"my_runtime_value": "a"}}))
+print(graph.invoke({}, context={"my_runtime_value": "a"}))
 # highlight-next-line
-print(graph.invoke({}, {"configurable": {"my_runtime_value": "b"}}))
+print(graph.invoke({}, context={"my_runtime_value": "b"}))
 ```
 ```
 {'my_state_value': 1}
@@ -560,27 +560,28 @@ print(graph.invoke({}, {"configurable": {"my_runtime_value": "b"}}))
     Below we demonstrate a practical example in which we configure what LLM to use at runtime. We will use both OpenAI and Anthropic models.
 
     ```python
+    from dataclasses import dataclass
+
     from langchain.chat_models import init_chat_model
-    from langchain_core.runnables import RunnableConfig
-    from langgraph.graph import MessagesState
-    from langgraph.graph import END, StateGraph, START
+    from langgraph.graph import MessagesState, END, StateGraph, START
+    from langgraph.runtime import Runtime
     from typing_extensions import TypedDict
 
-    class ConfigSchema(TypedDict):
-        model: str
+    @dataclass
+    class ContextSchema:
+        model_provider: str = "anthropic"
 
     MODELS = {
         "anthropic": init_chat_model("anthropic:claude-3-5-haiku-latest"),
         "openai": init_chat_model("openai:gpt-4.1-mini"),
     }
 
-    def call_model(state: MessagesState, config: RunnableConfig):
-        model = config["configurable"].get("model", "anthropic")
-        model = MODELS[model]
+    def call_model(state: MessagesState, runtime: Runtime[ContextSchema]):
+        model = MODELS[runtime.context.model_provider]
         response = model.invoke(state["messages"])
         return {"messages": [response]}
 
-    builder = StateGraph(MessagesState, config_schema=ConfigSchema)
+    builder = StateGraph(MessagesState, context_schema=ContextSchema)
     builder.add_node("model", call_model)
     builder.add_edge(START, "model")
     builder.add_edge("model", END)
@@ -592,8 +593,7 @@ print(graph.invoke({}, {"configurable": {"my_runtime_value": "b"}}))
     # With no configuration, uses default (Anthropic)
     response_1 = graph.invoke({"messages": [input_message]})["messages"][-1]
     # Or, can set OpenAI
-    config = {"configurable": {"model": "openai"}}
-    response_2 = graph.invoke({"messages": [input_message]}, config=config)["messages"][-1]
+    response_2 = graph.invoke({"messages": [input_message]}, context={"model_provider": "openai"})["messages"][-1]
 
     print(response_1.response_metadata["model_name"])
     print(response_2.response_metadata["model_name"])
@@ -607,32 +607,33 @@ print(graph.invoke({}, {"configurable": {"my_runtime_value": "b"}}))
     Below we demonstrate a practical example in which we configure two parameters: the LLM and system message to use at runtime.
 
     ```python
+    from dataclasses import dataclass
     from typing import Optional
     from langchain.chat_models import init_chat_model
     from langchain_core.messages import SystemMessage
-    from langchain_core.runnables import RunnableConfig
     from langgraph.graph import END, MessagesState, StateGraph, START
+    from langgraph.runtime import Runtime
     from typing_extensions import TypedDict
 
-    class ConfigSchema(TypedDict):
-        model: Optional[str]
-        system_message: Optional[str]
+    @dataclass
+    class ContextSchema:
+        model_provider: str = "anthropic"
+        system_message: str | None = None
 
     MODELS = {
         "anthropic": init_chat_model("anthropic:claude-3-5-haiku-latest"),
         "openai": init_chat_model("openai:gpt-4.1-mini"),
     }
 
-    def call_model(state: MessagesState, config: RunnableConfig):
-        model = config["configurable"].get("model", "anthropic")
-        model = MODELS[model]
+    def call_model(state: MessagesState, runtime: Runtime[ContextSchema]):
+        model = MODELS[runtime.context.model_provider]
         messages = state["messages"]
-        if system_message := config["configurable"].get("system_message"):
+        if (system_message := runtime.context.system_message):
             messages = [SystemMessage(system_message)] + messages
         response = model.invoke(messages)
         return {"messages": [response]}
 
-    builder = StateGraph(MessagesState, config_schema=ConfigSchema)
+    builder = StateGraph(MessagesState, context_schema=ContextSchema)
     builder.add_node("model", call_model)
     builder.add_edge(START, "model")
     builder.add_edge("model", END)
@@ -641,8 +642,7 @@ print(graph.invoke({}, {"configurable": {"my_runtime_value": "b"}}))
 
     # Usage
     input_message = {"role": "user", "content": "hi"}
-    config = {"configurable": {"model": "openai", "system_message": "Respond in Italian."}}
-    response = graph.invoke({"messages": [input_message]}, config)
+    response = graph.invoke({"messages": [input_message]}, context={"model_provider": "openai", "system_message": "Respond in Italian."})
     for message in response["messages"]:
         message.pretty_print()
     ```
@@ -1152,12 +1152,13 @@ LangGraph supports map-reduce and other advanced branching patterns using the Se
 ```python
 from langgraph.graph import StateGraph, START, END
 from langgraph.types import Send
-from typing_extensions import TypedDict
+from typing_extensions import TypedDict, Annotated
+import operator
 
 class OverallState(TypedDict):
     topic: str
     subjects: list[str]
-    jokes: list[str]
+    jokes: Annotated[list[str], operator.add]
     best_selected_joke: str
 
 def generate_topics(state: OverallState):
@@ -1566,9 +1567,9 @@ class State(TypedDict):
 
 def node_a(state: State) -> Command[Literal["node_b", "node_c"]]:
     print("Called A")
-    value = random.choice(["a", "b"])
+    value = random.choice(["b", "c"])
     # this is a replacement for a conditional edge function
-    if value == "a":
+    if value == "b":
         goto = "node_b"
     else:
         goto = "node_c"

docs/docs/how-tos/human_in_the_loop/add-human-in-the-loop.md

@@ -54,13 +54,7 @@ graph = graph_builder.compile(checkpointer=checkpointer) # (4)!
 config = {"configurable": {"thread_id": "some_id"}}
 result = graph.invoke({"some_text": "original text"}, config=config) # (5)!
 print(result['__interrupt__']) # (6)!
-# > [
-# >    Interrupt(
-# >       value={'text_to_revise': 'original text'}, 
-# >       resumable=True,
-# >       ns=['human_node:6ce9e64f-edef-fe5d-f7dc-511fa9526960']
-# >    )
-# > ] 
+# > [Interrupt(value={'text_to_revise': 'original text'}, id='a0d9dd40440ac7be2720dc5c20858627')]
 
 # highlight-next-line
 print(graph.invoke(Command(resume="Edited text"), config=config)) # (7)!
@@ -80,25 +74,27 @@ print(graph.invoke(Command(resume="Edited text"), config=config)) # (7)!
     ```python
     from typing import TypedDict
     import uuid
-
     from langgraph.checkpoint.memory import InMemorySaver
     from langgraph.constants import START
     from langgraph.graph import StateGraph
+
     # highlight-next-line
     from langgraph.types import interrupt, Command
 
+
     class State(TypedDict):
         some_text: str
 
+
     def human_node(state: State):
         # highlight-next-line
-        value = interrupt( # (1)!
+        value = interrupt(  # (1)!
             {
-                "text_to_revise": state["some_text"] # (2)!
+                "text_to_revise": state["some_text"]  # (2)!
             }
         )
         return {
-            "some_text": value # (3)!
+            "some_text": value  # (3)!
         }
 
 
@@ -106,25 +102,15 @@ print(graph.invoke(Command(resume="Edited text"), config=config)) # (7)!
     graph_builder = StateGraph(State)
     graph_builder.add_node("human_node", human_node)
     graph_builder.add_edge(START, "human_node")
-
-    checkpointer = InMemorySaver() # (4)!
-
+    checkpointer = InMemorySaver()  # (4)!
     graph = graph_builder.compile(checkpointer=checkpointer)
-
     # Pass a thread ID to the graph to run it.
     config = {"configurable": {"thread_id": uuid.uuid4()}}
-
     # Run the graph until the interrupt is hit.
-    result = graph.invoke({"some_text": "original text"}, config=config) # (5)!
+    result = graph.invoke({"some_text": "original text"}, config=config)  # (5)!
 
-    print(result['__interrupt__']) # (6)!
-    # > [
-    # >    Interrupt(
-    # >       value={'text_to_revise': 'original text'}, 
-    # >       resumable=True,
-    # >       ns=['human_node:6ce9e64f-edef-fe5d-f7dc-511fa9526960']
-    # >    )
-    # > ] 
+    print(result["__interrupt__"])  # (6)!
+    # > [Interrupt(value={'text_to_revise': 'original text'}, id='6d7c4048049254c83195429a3659661d')]
 
     # highlight-next-line
     print(graph.invoke(Command(resume="Edited text"), config=config)) # (7)!
@@ -167,7 +153,7 @@ For example, once your graph has been interrupted (multiple times, theoretically
 
 ```python
 resume_map = {
-    i.interrupt_id: f"human input for prompt {i.value}"
+    i.id: f"human input for prompt {i.value}"
     for i in parent.get_state(thread_config).interrupts
 }
 
@@ -226,7 +212,7 @@ graph.invoke(Command(resume=True), config=thread_config)
     from langgraph.constants import START, END
     from langgraph.graph import StateGraph
     from langgraph.types import interrupt, Command
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
     # Define the shared graph state
     class State(TypedDict):
@@ -271,7 +257,7 @@ graph.invoke(Command(resume=True), config=thread_config)
     builder.add_edge("approved_path", END)
     builder.add_edge("rejected_path", END)
 
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
     graph = builder.compile(checkpointer=checkpointer)
 
     # Run until interrupt
@@ -339,7 +325,7 @@ graph.invoke(
     from langgraph.constants import START, END
     from langgraph.graph import StateGraph
     from langgraph.types import interrupt, Command
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
     # Define the graph state
     class State(TypedDict):
@@ -378,7 +364,7 @@ graph.invoke(
     builder.add_edge("downstream_use", END)
 
     # Set up in-memory checkpointing for interrupt support
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
     graph = builder.compile(checkpointer=checkpointer)
 
     # Invoke the graph until it hits the interrupt
@@ -388,14 +374,15 @@ graph.invoke(
     # Output interrupt payload
     print(result["__interrupt__"])
     # Example output:
-    # Interrupt(
-    #   value={
-    #     'task': 'Please review and edit the generated summary if necessary.',
-    #     'generated_summary': 'The cat sat on the mat and looked at the stars.'
-    #   },
-    #   resumable=True,
-    #   ...
-    # )
+    # > [
+    # >     Interrupt(
+    # >         value={
+    # >             'task': 'Please review and edit the generated summary if necessary.',
+    # >             'generated_summary': 'The cat sat on the mat and looked at the stars.'
+    # >         }, 
+    # >         id='...'
+    # >     )
+    # > ]
 
     # Resume the graph with human-edited input
     edited_summary = "The cat lay on the rug, gazing peacefully at the night sky."
@@ -655,7 +642,7 @@ def human_node(state: State):
     from langgraph.constants import START, END
     from langgraph.graph import StateGraph
     from langgraph.types import interrupt, Command
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
     # Define graph state
     class State(TypedDict):
@@ -694,7 +681,7 @@ def human_node(state: State):
     builder.add_edge("report_age", END)
 
     # Create the graph with a memory checkpointer
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
     graph = builder.compile(checkpointer=checkpointer)
 
     # Run the graph until the first interrupt
@@ -951,7 +938,7 @@ def node_in_parent_graph(state: State):
     from langgraph.graph import StateGraph
     from langgraph.constants import START
     from langgraph.types import interrupt, Command
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
 
     class State(TypedDict):
@@ -977,7 +964,7 @@ def node_in_parent_graph(state: State):
         print(f"Got an answer of {answer}")
 
 
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
 
     subgraph_builder = StateGraph(State)
     subgraph_builder.add_node("some_node", node_in_subgraph)
@@ -1008,7 +995,7 @@ def node_in_parent_graph(state: State):
     builder.add_edge(START, "parent_node")
 
     # A checkpointer must be enabled for interrupts to work!
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
     graph = builder.compile(checkpointer=checkpointer)
 
     config = {
@@ -1032,7 +1019,7 @@ def node_in_parent_graph(state: State):
     Entered `parent_node` a total of 1 times
     Entered `node_in_subgraph` a total of 1 times
     Entered human_node in sub-graph a total of 1 times
-    {'__interrupt__': (Interrupt(value='what is your name?', resumable=True, ns=['parent_node:4c3a0248-21f0-1287-eacf-3002bc304db4', 'human_node:2fe86d52-6f70-2a3f-6b2f-b1eededd6348'], when='during'),)}
+    {'__interrupt__': (Interrupt(value='what is your name?', id='...'),)}
     --- Resuming ---
     Entered `parent_node` a total of 2 times
     Entered human_node in sub-graph a total of 2 times
@@ -1057,7 +1044,7 @@ To avoid issues, refrain from dynamically changing the node's structure between
     from langgraph.graph import StateGraph
     from langgraph.constants import START 
     from langgraph.types import interrupt, Command
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
 
     class State(TypedDict):
@@ -1091,7 +1078,7 @@ To avoid issues, refrain from dynamically changing the node's structure between
     builder.add_edge(START, "human_node")
 
     # A checkpointer must be enabled for interrupts to work!
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
     graph = builder.compile(checkpointer=checkpointer)
 
     config = {
@@ -1108,7 +1095,7 @@ To avoid issues, refrain from dynamically changing the node's structure between
     ```
 
     ```pycon
-    {'__interrupt__': (Interrupt(value='what is your name?', resumable=True, ns=['human_node:3a007ef9-c30d-c357-1ec1-86a1a70d8fba'], when='during'),)}
+    {'__interrupt__': (Interrupt(value='what is your name?', id='...'),)}
     Name: N/A. Age: John
     {'human_node': {'age': 'John', 'name': 'N/A'}}
     ```

docs/docs/how-tos/human_in_the_loop/wait-user-input.ipynb

@@ -121,7 +121,7 @@
     "\n",
     "# highlight-next-line\n",
     "from langgraph.types import Command, interrupt\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from IPython.display import Image, display\n",
     "\n",
     "\n",
@@ -157,7 +157,7 @@
     "builder.add_edge(\"step_3\", END)\n",
     "\n",
     "# Set up memory\n",
-    "memory = MemorySaver()\n",
+    "memory = InMemorySaver()\n",
     "\n",
     "# Add\n",
     "graph = builder.compile(checkpointer=memory)\n",
@@ -435,9 +435,9 @@
     "workflow.add_edge(\"ask_human\", \"agent\")\n",
     "\n",
     "# Set up memory\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "\n",
-    "memory = MemorySaver()\n",
+    "memory = InMemorySaver()\n",
     "\n",
     "# Finally, we compile it!\n",
     "# This compiles it into a LangChain Runnable,\n",

docs/docs/how-tos/multi-agent-multi-turn-convo-functional.ipynb

@@ -224,7 +224,7 @@
     "from langgraph.prebuilt import create_react_agent\n",
     "from langgraph.graph import add_messages\n",
     "from langgraph.func import entrypoint, task\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from langgraph.types import interrupt, Command\n",
     "\n",
     "model = ChatAnthropic(model=\"claude-3-5-sonnet-latest\")\n",
@@ -272,7 +272,7 @@
     "    return response[\"messages\"]\n",
     "\n",
     "\n",
-    "checkpointer = MemorySaver()\n",
+    "checkpointer = InMemorySaver()\n",
     "\n",
     "\n",
     "def string_to_uuid(input_string):\n",

docs/docs/how-tos/multi_agent.md

@@ -375,7 +375,7 @@ def agent(state) -> Command[Literal["agent", "another_agent", "human"]]:
     from langgraph.graph import MessagesState, StateGraph, START
     from langgraph.prebuilt import create_react_agent, InjectedState
     from langgraph.types import Command, interrupt
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
     
     
     model = ChatAnthropic(model="claude-3-5-sonnet-latest")
@@ -467,7 +467,7 @@ def agent(state) -> Command[Literal["agent", "another_agent", "human"]]:
     builder.add_edge(START, "travel_advisor")
     
     
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
     graph = builder.compile(checkpointer=checkpointer)
     ```
     

docs/docs/how-tos/persistence-functional.ipynb

@@ -28,9 +28,9 @@
     "1. Create an instance of a checkpointer:\n",
     "\n",
     "    ```python\n",
-    "    from langgraph.checkpoint.memory import MemorySaver\n",
+    "    from langgraph.checkpoint.memory import InMemorySaver\n",
     "    \n",
-    "    checkpointer = MemorySaver()       \n",
+    "    checkpointer = InMemorySaver()       \n",
     "    ```\n",
     "\n",
     "2. Pass `checkpointer` instance to the `entrypoint()` decorator:\n",
@@ -184,7 +184,7 @@
     "from langchain_core.messages import BaseMessage\n",
     "from langgraph.graph import add_messages\n",
     "from langgraph.func import entrypoint, task\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "\n",
     "\n",
     "@task\n",
@@ -193,7 +193,7 @@
     "    return response\n",
     "\n",
     "\n",
-    "checkpointer = MemorySaver()\n",
+    "checkpointer = InMemorySaver()\n",
     "\n",
     "\n",
     "@entrypoint(checkpointer=checkpointer)\n",

docs/docs/how-tos/react-agent-from-scratch-functional.ipynb

@@ -261,7 +261,7 @@
     "\n",
     "To add thread-level persistence to our agent:\n",
     "\n",
-    "1. Select a [checkpointer](../../concepts/persistence#checkpointer-libraries): here we will use [MemorySaver](../../reference/checkpoints/#langgraph.checkpoint.memory.MemorySaver), a simple in-memory checkpointer.\n",
+    "1. Select a [checkpointer](../../concepts/persistence#checkpointer-libraries): here we will use [InMemorySaver](../../reference/checkpoints/#langgraph.checkpoint.memory.InMemorySaver), a simple in-memory checkpointer.\n",
     "2. Update our entrypoint to accept the previous messages state as a second argument. Here, we simply append the message updates to the previous sequence of messages.\n",
     "3. Choose which values will be returned from the workflow and which will be saved by the checkpointer as `previous` using `entrypoint.final` (optional)"
    ]
@@ -272,10 +272,10 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "\n",
     "# highlight-next-line\n",
-    "checkpointer = MemorySaver()\n",
+    "checkpointer = InMemorySaver()\n",
     "\n",
     "\n",
     "# highlight-next-line\n",

docs/docs/how-tos/use-functional-api.md

@@ -26,7 +26,7 @@ my_workflow.invoke({"value": 1, "another_value": 2})
     ```python
     import uuid
     from langgraph.func import entrypoint, task
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
     # Task that checks if a number is even
     @task
@@ -39,7 +39,7 @@ my_workflow.invoke({"value": 1, "another_value": 2})
         return "The number is even." if is_even else "The number is odd."
 
     # Create a checkpointer for persistence
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
 
     @entrypoint(checkpointer=checkpointer)
     def workflow(inputs: dict) -> str:
@@ -63,7 +63,7 @@ my_workflow.invoke({"value": 1, "another_value": 2})
     import uuid
     from langchain.chat_models import init_chat_model
     from langgraph.func import entrypoint, task
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
     llm = init_chat_model('openai:gpt-3.5-turbo')
 
@@ -77,7 +77,7 @@ my_workflow.invoke({"value": 1, "another_value": 2})
         ]).content
 
     # Create a checkpointer for persistence
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
 
     @entrypoint(checkpointer=checkpointer)
     def workflow(topic: str) -> str:
@@ -114,7 +114,7 @@ def graph(numbers: list[int]) -> list[str]:
     import uuid
     from langchain.chat_models import init_chat_model
     from langgraph.func import entrypoint, task
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
     # Initialize the LLM model
     llm = init_chat_model("openai:gpt-3.5-turbo")
@@ -129,7 +129,7 @@ def graph(numbers: list[int]) -> list[str]:
         return response.content
 
     # Create a checkpointer for persistence
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
 
     @entrypoint(checkpointer=checkpointer)
     def workflow(topics: list[str]) -> str:
@@ -176,7 +176,7 @@ def some_workflow(some_input: dict) -> int:
     import uuid
     from typing import TypedDict
     from langgraph.func import entrypoint
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
     from langgraph.graph import StateGraph
 
     # Define the shared state type
@@ -194,7 +194,7 @@ def some_workflow(some_input: dict) -> int:
     graph = builder.compile()
 
     # Define the functional API workflow
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
 
     @entrypoint(checkpointer=checkpointer)
     def workflow(x: int) -> dict:
@@ -227,10 +227,10 @@ def my_workflow(inputs: dict) -> int:
     ```python
     import uuid
     from langgraph.func import entrypoint
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
     # Initialize a checkpointer
-    checkpointer = MemorySaver()
+    checkpointer = InMemorySaver()
 
     # A reusable sub-workflow that multiplies a number
     @entrypoint()
@@ -258,10 +258,10 @@ Example of using the streaming API to stream both updates and custom data.
 
 ```python
 from langgraph.func import entrypoint
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.config import get_stream_writer # (1)!
 
-checkpointer = MemorySaver()
+checkpointer = InMemorySaver()
 
 @entrypoint(checkpointer=checkpointer)
 def main(inputs: dict) -> int:
@@ -316,7 +316,7 @@ for mode, chunk in main.stream( # (5)!
 ## Retry policy
 
 ```python
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.func import entrypoint, task
 from langgraph.types import RetryPolicy
 
@@ -337,7 +337,7 @@ def get_info():
         raise ValueError('Failure')
     return "OK"
 
-checkpointer = MemorySaver()
+checkpointer = InMemorySaver()
 
 @entrypoint(checkpointer=checkpointer)
 def main(inputs, writer):
@@ -392,7 +392,7 @@ for chunk in main.stream({"x": 5}, stream_mode="updates"):
 
 ```python
 import time
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.func import entrypoint, task
 from langgraph.types import StreamWriter
 
@@ -414,7 +414,7 @@ def get_info():
     return "OK"
 
 # Initialize an in-memory checkpointer for persistence
-checkpointer = MemorySaver()
+checkpointer = InMemorySaver()
 
 @task
 def slow_task():
@@ -504,9 +504,9 @@ def step_3(input_query):
 We can now compose these tasks in an [entrypoint](../concepts/functional_api.md#entrypoint):
 
 ```python
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 
-checkpointer = MemorySaver()
+checkpointer = InMemorySaver()
 
 
 @entrypoint(checkpointer=checkpointer)
@@ -577,12 +577,12 @@ def review_tool_call(tool_call: ToolCall) -> Union[ToolCall, ToolMessage]:
 We can now update our [entrypoint](../concepts/functional_api.md#entrypoint) to review the generated tool calls. If a tool call is accepted or revised, we execute in the same way as before. Otherwise, we just append the `ToolMessage` supplied by the human. The results of prior tasks — in this case the initial model call — are persisted, so that they are not run again following the `interrupt`.
 
 ```python
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.graph.message import add_messages
 from langgraph.types import Command, interrupt
 
 
-checkpointer = MemorySaver()
+checkpointer = InMemorySaver()
 
 
 @entrypoint(checkpointer=checkpointer)
@@ -757,9 +757,9 @@ Use `entrypoint.final` to decouple what is returned to the caller from what is p
 ```python
 from typing import Optional
 from langgraph.func import entrypoint
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 
-checkpointer = MemorySaver()
+checkpointer = InMemorySaver()
 
 @entrypoint(checkpointer=checkpointer)
 def accumulate(n: int, *, previous: Optional[int]) -> entrypoint.final[int, int]:
@@ -777,14 +777,14 @@ print(accumulate.invoke(3, config=config))  # 3
 
 ### Chatbot example
 
-An example of a simple chatbot using the functional API and the `MemorySaver` checkpointer.
+An example of a simple chatbot using the functional API and the `InMemorySaver` checkpointer.
 The bot is able to remember the previous conversation and continue from where it left off.
 
 ```python
 from langchain_core.messages import BaseMessage
 from langgraph.graph import add_messages
 from langgraph.func import entrypoint, task
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langchain_anthropic import ChatAnthropic
 
 model = ChatAnthropic(model="claude-3-5-sonnet-latest")
@@ -794,7 +794,7 @@ def call_model(messages: list[BaseMessage]):
     response = model.invoke(messages)
     return response
 
-checkpointer = MemorySaver()
+checkpointer = InMemorySaver()
 
 @entrypoint(checkpointer=checkpointer)
 def workflow(inputs: list[BaseMessage], *, previous: list[BaseMessage]):

docs/docs/reference/constants.md

@@ -2,5 +2,6 @@
     options:
       members:
         - TAG_HIDDEN
+        - TAG_NOSTREAM
         - START
-        - END
+        - END

docs/docs/tutorials/chatbots/information-gather-prompting.ipynb

@@ -256,7 +256,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from langgraph.graph import StateGraph, START\n",
     "from langgraph.graph.message import add_messages\n",
     "from typing import Annotated\n",
@@ -267,7 +267,7 @@
     "    messages: Annotated[list, add_messages]\n",
     "\n",
     "\n",
-    "memory = MemorySaver()\n",
+    "memory = InMemorySaver()\n",
     "workflow = StateGraph(State)\n",
     "workflow.add_node(\"info\", info_chain)\n",
     "workflow.add_node(\"prompt\", prompt_gen_chain)\n",

docs/docs/tutorials/customer-support/customer-support.ipynb

@@ -1124,7 +1124,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from langgraph.graph import END, StateGraph, START\n",
     "from langgraph.prebuilt import tools_condition\n",
     "\n",
@@ -1144,7 +1144,7 @@
     "\n",
     "# The checkpointer lets the graph persist its state\n",
     "# this is a complete memory for the entire graph.\n",
-    "memory = MemorySaver()\n",
+    "memory = InMemorySaver()\n",
     "part_1_graph = builder.compile(checkpointer=memory)"
    ]
   },
@@ -1943,7 +1943,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from langgraph.graph import StateGraph\n",
     "from langgraph.prebuilt import tools_condition\n",
     "\n",
@@ -1967,7 +1967,7 @@
     ")\n",
     "builder.add_edge(\"tools\", \"assistant\")\n",
     "\n",
-    "memory = MemorySaver()\n",
+    "memory = InMemorySaver()\n",
     "part_2_graph = builder.compile(\n",
     "    checkpointer=memory,\n",
     "    # NEW: The graph will always halt before executing the \"tools\" node.\n",
@@ -2532,7 +2532,7 @@
    "source": [
     "from typing import Literal\n",
     "\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from langgraph.graph import StateGraph\n",
     "from langgraph.prebuilt import tools_condition\n",
     "\n",
@@ -2576,7 +2576,7 @@
     "builder.add_edge(\"safe_tools\", \"assistant\")\n",
     "builder.add_edge(\"sensitive_tools\", \"assistant\")\n",
     "\n",
-    "memory = MemorySaver()\n",
+    "memory = InMemorySaver()\n",
     "part_3_graph = builder.compile(\n",
     "    checkpointer=memory,\n",
     "    # NEW: The graph will always halt before executing the \"tools\" node.\n",
@@ -3477,7 +3477,7 @@
    "source": [
     "from typing import Literal\n",
     "\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from langgraph.graph import StateGraph\n",
     "from langgraph.prebuilt import tools_condition\n",
     "\n",
@@ -3841,7 +3841,7 @@
     "builder.add_conditional_edges(\"fetch_user_info\", route_to_workflow)\n",
     "\n",
     "# Compile graph\n",
-    "memory = MemorySaver()\n",
+    "memory = InMemorySaver()\n",
     "part_4_graph = builder.compile(\n",
     "    checkpointer=memory,\n",
     "    # Let the user approve or deny the use of sensitive tools\n",

docs/docs/tutorials/get-started/3-add-memory.md

@@ -10,14 +10,14 @@ We will see later that **checkpointing** is _much_ more powerful than simple cha
 
     This tutorial builds on [Add tools](./2-add-tools.md).
 
-## 1. Create a `MemorySaver` checkpointer
+## 1. Create a `InMemorySaver` checkpointer
 
-Create a `MemorySaver` checkpointer:
+Create a `InMemorySaver` checkpointer:
 
 ``` python
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 
-memory = MemorySaver()
+memory = InMemorySaver()
 ```
 
 This is in-memory checkpointer, which is convenient for the tutorial. However, in a production application, you would likely change this to use `SqliteSaver` or `PostgresSaver` and connect a database.
@@ -172,7 +172,7 @@ from langchain_tavily import TavilySearch
 from langchain_core.messages import BaseMessage
 from typing_extensions import TypedDict
 
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.graph import StateGraph
 from langgraph.graph.message import add_messages
 from langgraph.prebuilt import ToolNode, tools_condition
@@ -200,7 +200,7 @@ graph_builder.add_conditional_edges(
 )
 graph_builder.add_edge("tools", "chatbot")
 graph_builder.set_entry_point("chatbot")
-memory = MemorySaver()
+memory = InMemorySaver()
 graph = graph_builder.compile(checkpointer=memory)
 ```
 

docs/docs/tutorials/get-started/4-human-in-the-loop.md

@@ -33,7 +33,7 @@ from langchain_tavily import TavilySearch
 from langchain_core.tools import tool
 from typing_extensions import TypedDict
 
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.graph import StateGraph, START, END
 from langgraph.graph.message import add_messages
 from langgraph.prebuilt import ToolNode, tools_condition
@@ -85,7 +85,7 @@ graph_builder.add_edge(START, "chatbot")
 We compile the graph with a checkpointer, as before:
 
 ```python
-memory = MemorySaver()
+memory = InMemorySaver()
 
 graph = graph_builder.compile(checkpointer=memory)
 ```
@@ -230,7 +230,7 @@ from langchain_tavily import TavilySearch
 from langchain_core.tools import tool
 from typing_extensions import TypedDict
 
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.graph import StateGraph, START, END
 from langgraph.graph.message import add_messages
 from langgraph.prebuilt import ToolNode, tools_condition
@@ -268,7 +268,7 @@ graph_builder.add_conditional_edges(
 graph_builder.add_edge("tools", "chatbot")
 graph_builder.add_edge(START, "chatbot")
 
-memory = MemorySaver()
+memory = InMemorySaver()
 graph = graph_builder.compile(checkpointer=memory)
 ```
 

docs/docs/tutorials/get-started/5-customize-state.md

@@ -239,7 +239,7 @@ from langchain_core.messages import ToolMessage
 from langchain_core.tools import InjectedToolCallId, tool
 from typing_extensions import TypedDict
 
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.graph import StateGraph, START, END
 from langgraph.graph.message import add_messages
 from langgraph.prebuilt import ToolNode, tools_condition
@@ -301,7 +301,7 @@ graph_builder.add_conditional_edges(
 graph_builder.add_edge("tools", "chatbot")
 graph_builder.add_edge(START, "chatbot")
 
-memory = MemorySaver()
+memory = InMemorySaver()
 graph = graph_builder.compile(checkpointer=memory)
 ```
 

docs/docs/tutorials/get-started/6-time-travel.md

@@ -31,7 +31,7 @@ from langchain_tavily import TavilySearch
 from langchain_core.messages import BaseMessage
 from typing_extensions import TypedDict
 
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.graph import StateGraph, START, END
 from langgraph.graph.message import add_messages
 from langgraph.prebuilt import ToolNode, tools_condition
@@ -60,7 +60,7 @@ graph_builder.add_conditional_edges(
 graph_builder.add_edge("tools", "chatbot")
 graph_builder.add_edge(START, "chatbot")
 
-memory = MemorySaver()
+memory = InMemorySaver()
 graph = graph_builder.compile(checkpointer=memory)
 ```
 

docs/docs/tutorials/langgraph-platform/local-server.md

@@ -12,9 +12,9 @@ Before you begin, ensure you have the following:
 
 === "Python server"
 
-    ```shell
-    # Python >= 3.11 is required.
+    Python >= 3.11 is required.
 
+    ```shell
     pip install --upgrade "langgraph-cli[inmem]"
     ```
 

docs/docs/tutorials/reflection/reflection.ipynb

@@ -322,7 +322,7 @@
     "from typing import Annotated, List, Sequence\n",
     "from langgraph.graph import END, StateGraph, START\n",
     "from langgraph.graph.message import add_messages\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from typing_extensions import TypedDict\n",
     "\n",
     "\n",
@@ -361,7 +361,7 @@
     "\n",
     "builder.add_conditional_edges(\"generate\", should_continue)\n",
     "builder.add_edge(\"reflect\", \"generate\")\n",
-    "memory = MemorySaver()\n",
+    "memory = InMemorySaver()\n",
     "graph = builder.compile(checkpointer=memory)"
    ]
   },

docs/docs/tutorials/tot/tot.ipynb

@@ -272,7 +272,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -280,10 +280,10 @@
     "from typing import Optional, Dict, Any\n",
     "from typing_extensions import Annotated, TypedDict\n",
     "from langgraph.graph import StateGraph\n",
+    "from langgraph.runtime import Runtime\n",
     "\n",
-    "from langchain_core.runnables import RunnableConfig\n",
-    "from langgraph.constants import Send\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
+    "from langgraph.types import Send\n",
     "\n",
     "\n",
     "def update_candidates(\n",
@@ -307,22 +307,27 @@
     "    depth: Annotated[int, operator.add]\n",
     "\n",
     "\n",
-    "class Configuration(TypedDict, total=False):\n",
+    "class Context(TypedDict, total=False):\n",
     "    max_depth: int\n",
     "    threshold: float\n",
     "    k: int\n",
     "    beam_size: int\n",
     "\n",
     "\n",
-    "def _ensure_configurable(config: RunnableConfig) -> Configuration:\n",
+    "class EnsuredContext(TypedDict):\n",
+    "    max_depth: int\n",
+    "    threshold: float\n",
+    "    k: int\n",
+    "    beam_size: int\n",
+    "\n",
+    "\n",
+    "def _ensure_context(ctx: Context) -> EnsuredContext:\n",
     "    \"\"\"Get params that configure the search algorithm.\"\"\"\n",
-    "    configurable = config.get(\"configurable\", {})\n",
     "    return {\n",
-    "        **configurable,\n",
-    "        \"max_depth\": configurable.get(\"max_depth\", 10),\n",
-    "        \"threshold\": config.get(\"threshold\", 0.9),\n",
-    "        \"k\": configurable.get(\"k\", 5),\n",
-    "        \"beam_size\": configurable.get(\"beam_size\", 3),\n",
+    "        \"max_depth\": ctx.get(\"max_depth\", 10),\n",
+    "        \"threshold\": ctx.get(\"threshold\", 0.9),\n",
+    "        \"k\": ctx.get(\"k\", 5),\n",
+    "        \"beam_size\": ctx.get(\"beam_size\", 3),\n",
     "    }\n",
     "\n",
     "\n",
@@ -330,9 +335,11 @@
     "    seed: Optional[Candidate]\n",
     "\n",
     "\n",
-    "def expand(state: ExpansionState, *, config: RunnableConfig) -> Dict[str, List[str]]:\n",
+    "def expand(\n",
+    "    state: ExpansionState, *, runtime: Runtime[Context]\n",
+    ") -> Dict[str, List[Candidate]]:\n",
     "    \"\"\"Generate the next state.\"\"\"\n",
-    "    configurable = _ensure_configurable(config)\n",
+    "    ctx = _ensure_context(runtime.context)\n",
     "    if not state.get(\"seed\"):\n",
     "        candidate_str = \"\"\n",
     "    else:\n",
@@ -342,9 +349,8 @@
     "            {\n",
     "                \"problem\": state[\"problem\"],\n",
     "                \"candidate\": candidate_str,\n",
-    "                \"k\": configurable[\"k\"],\n",
+    "                \"k\": ctx[\"k\"],\n",
     "            },\n",
-    "            config=config,\n",
     "        )\n",
     "    except Exception:\n",
     "        return {\"candidates\": []}\n",
@@ -354,7 +360,7 @@
     "    return {\"candidates\": new_candidates}\n",
     "\n",
     "\n",
-    "def score(state: ToTState) -> Dict[str, List[float]]:\n",
+    "def score(state: ToTState) -> Dict[str, Any]:\n",
     "    \"\"\"Evaluate the candidate generations.\"\"\"\n",
     "    candidates = state[\"candidates\"]\n",
     "    scored = []\n",
@@ -363,11 +369,9 @@
     "    return {\"scored_candidates\": scored, \"candidates\": \"clear\"}\n",
     "\n",
     "\n",
-    "def prune(\n",
-    "    state: ToTState, *, config: RunnableConfig\n",
-    ") -> Dict[str, List[Dict[str, Any]]]:\n",
+    "def prune(state: ToTState, *, runtime: Runtime[Context]) -> Dict[str, Any]:\n",
     "    scored_candidates = state[\"scored_candidates\"]\n",
-    "    beam_size = _ensure_configurable(config)[\"beam_size\"]\n",
+    "    beam_size = _ensure_context(runtime.context)[\"beam_size\"]\n",
     "    organized = sorted(\n",
     "        scored_candidates, key=lambda candidate: candidate[1], reverse=True\n",
     "    )\n",
@@ -383,11 +387,11 @@
     "\n",
     "\n",
     "def should_terminate(\n",
-    "    state: ToTState, config: RunnableConfig\n",
+    "    state: ToTState, runtime: Runtime[Context]\n",
     ") -> Union[Literal[\"__end__\"], Send]:\n",
-    "    configurable = _ensure_configurable(config)\n",
-    "    solved = state[\"candidates\"][0].score >= configurable[\"threshold\"]\n",
-    "    if solved or state[\"depth\"] >= configurable[\"max_depth\"]:\n",
+    "    ctx = _ensure_context(runtime.context)\n",
+    "    solved = state[\"candidates\"][0].score >= ctx[\"threshold\"]\n",
+    "    if solved or state[\"depth\"] >= ctx[\"max_depth\"]:\n",
     "        return \"__end__\"\n",
     "    return [\n",
     "        Send(\"expand\", {**state, \"somevalseed\": candidate})\n",
@@ -396,7 +400,7 @@
     "\n",
     "\n",
     "# Create the graph\n",
-    "builder = StateGraph(state_schema=ToTState, config_schema=Configuration)\n",
+    "builder = StateGraph(state_schema=ToTState, context_schema=Context)\n",
     "\n",
     "# Add nodes\n",
     "builder.add_node(expand)\n",
@@ -412,7 +416,7 @@
     "builder.add_edge(\"__start__\", \"expand\")\n",
     "\n",
     "# Compile the graph\n",
-    "graph = builder.compile(checkpointer=MemorySaver())"
+    "graph = builder.compile(checkpointer=InMemorySaver())"
    ]
   },
   {
@@ -467,13 +471,11 @@
     }
    ],
    "source": [
-    "config = {\n",
-    "    \"configurable\": {\n",
-    "        \"thread_id\": \"test_1\",\n",
-    "        \"depth\": 10,\n",
-    "    }\n",
-    "}\n",
-    "for step in graph.stream({\"problem\": puzzles[42]}, config):\n",
+    "for step in graph.stream(\n",
+    "    {\"problem\": puzzles[42]},\n",
+    "    config={\"configurable\": {\"thread_id\": \"test_1\"}},\n",
+    "    context={\"depth\": 10},\n",
+    "):\n",
     "    print(step)"
    ]
   },
@@ -491,7 +493,7 @@
     }
    ],
    "source": [
-    "final_state = graph.get_state(config)\n",
+    "final_state = graph.get_state({\"configurable\": {\"thread_id\": \"test_1\"}})\n",
     "winning_solution = final_state.values[\"candidates\"][0]\n",
     "search_depth = final_state.values[\"depth\"]\n",
     "if winning_solution[1] == 1:\n",

docs/docs/tutorials/usaco/usaco.ipynb

@@ -1029,7 +1029,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from langgraph.graph import END, StateGraph, START\n",
     "\n",
     "builder = StateGraph(State)\n",
@@ -1053,7 +1053,7 @@
     "builder.add_conditional_edges(\"evaluate\", control_edge, {END: END, \"solve\": \"solve\"})\n",
     "\n",
     "\n",
-    "checkpointer = MemorySaver()\n",
+    "checkpointer = InMemorySaver()\n",
     "graph = builder.compile(checkpointer=checkpointer)"
    ]
   },
@@ -1327,7 +1327,7 @@
    "outputs": [],
    "source": [
     "# This is all the same as before\n",
-    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from langgraph.checkpoint.memory import InMemorySaver\n",
     "from langgraph.graph import END, StateGraph, START\n",
     "\n",
     "builder = StateGraph(State)\n",
@@ -1353,7 +1353,7 @@
     "\n",
     "\n",
     "builder.add_conditional_edges(\"evaluate\", control_edge, {END: END, \"solve\": \"solve\"})\n",
-    "checkpointer = MemorySaver()"
+    "checkpointer = InMemorySaver()"
    ]
   },
   {

docs/uv.lock

@@ -15,16 +15,16 @@ name = "ag2"
 version = "0.9.6"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "anyio" },
-    { name = "asyncer" },
-    { name = "diskcache" },
-    { name = "docker" },
-    { name = "httpx" },
-    { name = "packaging" },
-    { name = "pydantic" },
-    { name = "python-dotenv" },
-    { name = "termcolor" },
-    { name = "tiktoken" },
+    { name = "anyio", marker = "python_full_version < '3.13'" },
+    { name = "asyncer", marker = "python_full_version < '3.13'" },
+    { name = "diskcache", marker = "python_full_version < '3.13'" },
+    { name = "docker", marker = "python_full_version < '3.13'" },
+    { name = "httpx", marker = "python_full_version < '3.13'" },
+    { name = "packaging", marker = "python_full_version < '3.13'" },
+    { name = "pydantic", marker = "python_full_version < '3.13'" },
+    { name = "python-dotenv", marker = "python_full_version < '3.13'" },
+    { name = "termcolor", marker = "python_full_version < '3.13'" },
+    { name = "tiktoken", marker = "python_full_version < '3.13'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/ee/15/edfbbf217e19ea647225b3ab72a6e3755d2677665f1a7f8e5108da3feabd/ag2-0.9.6.tar.gz", hash = "sha256:d6f7812b1a49654d14113fa3c13ccb593115dee1193744ca428d7178d2b32090", size = 3356270, upload-time = "2025-07-08T14:56:21.63Z" }
 wheels = [
@@ -267,7 +267,7 @@ name = "asyncer"
 version = "0.0.8"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "anyio" },
+    { name = "anyio", marker = "python_full_version < '3.13'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/ff/67/7ea59c3e69eaeee42e7fc91a5be67ca5849c8979acac2b920249760c6af2/asyncer-0.0.8.tar.gz", hash = "sha256:a589d980f57e20efb07ed91d0dbe67f1d2fd343e7142c66d3a099f05c620739c", size = 18217, upload-time = "2024-08-24T23:15:36.449Z" }
 wheels = [
@@ -288,7 +288,7 @@ name = "autogen"
 version = "0.9.6"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "ag2" },
+    { name = "ag2", marker = "python_full_version < '3.13'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/67/b9/dc958031b7e08ee50e3d40f5991f4c0bc21538df8d53aa3e9a9f2e2f7818/autogen-0.9.6.tar.gz", hash = "sha256:dc2efbeef61002608983afb120e62f8a109815eb741bcbc9ef398dcff7424a30", size = 43422, upload-time = "2025-07-08T14:56:17.6Z" }
 wheels = [
@@ -914,9 +914,9 @@ name = "docker"
 version = "7.1.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "pywin32", marker = "sys_platform == 'win32'" },
-    { name = "requests" },
-    { name = "urllib3" },
+    { name = "pywin32", marker = "python_full_version < '3.13' and sys_platform == 'win32'" },
+    { name = "requests", marker = "python_full_version < '3.13'" },
+    { name = "urllib3", marker = "python_full_version < '3.13'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/91/9b/4a2ea29aeba62471211598dac5d96825bb49348fa07e906ea930394a83ce/docker-7.1.0.tar.gz", hash = "sha256:ad8c70e6e3f8926cb8a92619b832b4ea5299e2831c14284663184e200546fa6c", size = 117834, upload-time = "2024-05-23T11:13:57.216Z" }
 wheels = [
@@ -2337,7 +2337,7 @@ wheels = [
 
 [[package]]
 name = "langgraph"
-version = "0.5.2"
+version = "0.6.0a1"
 source = { editable = "../libs/langgraph" }
 dependencies = [
     { name = "langchain-core" },
@@ -2365,7 +2365,7 @@ dev = [
     { name = "langgraph-checkpoint", editable = "../libs/checkpoint" },
     { name = "langgraph-checkpoint-postgres", editable = "../libs/checkpoint-postgres" },
     { name = "langgraph-checkpoint-sqlite", editable = "../libs/checkpoint-sqlite" },
-    { name = "langgraph-cli", extras = ["inmem"] },
+    { name = "langgraph-cli", extras = ["inmem"], editable = "../libs/cli" },
     { name = "langgraph-prebuilt", editable = "../libs/prebuilt" },
     { name = "langgraph-sdk", editable = "../libs/sdk-py" },
     { name = "mypy" },
@@ -2388,7 +2388,7 @@ dev = [
 
 [[package]]
 name = "langgraph-checkpoint"
-version = "2.1.0"
+version = "2.1.1"
 source = { editable = "../libs/checkpoint" }
 dependencies = [
     { name = "langchain-core" },
@@ -2433,7 +2433,7 @@ wheels = [
 
 [[package]]
 name = "langgraph-checkpoint-postgres"
-version = "2.0.21"
+version = "2.0.23"
 source = { editable = "../libs/checkpoint-postgres" }
 dependencies = [
     { name = "langgraph-checkpoint" },
@@ -2674,7 +2674,7 @@ dev = [
 
 [[package]]
 name = "langgraph-sdk"
-version = "0.1.72"
+version = "0.2.0a1"
 source = { editable = "../libs/sdk-py" }
 dependencies = [
     { name = "httpx" },

examples/code_assistant/langgraph_code_assistant_mistral.ipynb

@@ -154,7 +154,7 @@
    "id": "2dff2209-44c7-4e2c-b607-ba6675f9e45f",
    "metadata": {},
    "outputs": [],
-   "source": ["from langgraph.checkpoint.memory import MemorySaver\nfrom langgraph.graph import END, StateGraph, START\n\nbuilder = StateGraph(GraphState)\n\n# Define the nodes\nbuilder.add_node(\"generate\", generate)  # generation solution\nbuilder.add_node(\"check_code\", code_check)  # check code\n\n# Build graph\nbuilder.add_edge(START, \"generate\")\nbuilder.add_edge(\"generate\", \"check_code\")\nbuilder.add_conditional_edges(\n    \"check_code\",\n    decide_to_finish,\n    {\n        \"end\": END,\n        \"generate\": \"generate\",\n    },\n)\n\nmemory = MemorySaver()\ngraph = builder.compile(checkpointer=memory)"]
+   "source": ["from langgraph.checkpoint.memory import InMemorySaver\nfrom langgraph.graph import END, StateGraph, START\n\nbuilder = StateGraph(GraphState)\n\n# Define the nodes\nbuilder.add_node(\"generate\", generate)  # generation solution\nbuilder.add_node(\"check_code\", code_check)  # check code\n\n# Build graph\nbuilder.add_edge(START, \"generate\")\nbuilder.add_edge(\"generate\", \"check_code\")\nbuilder.add_conditional_edges(\n    \"check_code\",\n    decide_to_finish,\n    {\n        \"end\": END,\n        \"generate\": \"generate\",\n    },\n)\n\nmemory = InMemorySaver()\ngraph = builder.compile(checkpointer=memory)"]
   },
   {
    "cell_type": "code",

libs/checkpoint-postgres/langgraph/checkpoint/postgres/__init__.py

@@ -284,10 +284,7 @@ class PostgresSaver(BasePostgresSaver):
         configurable = config["configurable"].copy()
         thread_id = configurable.pop("thread_id")
         checkpoint_ns = configurable.pop("checkpoint_ns")
-        checkpoint_id = configurable.pop(
-            "checkpoint_id", configurable.pop("thread_ts", None)
-        )
-
+        checkpoint_id = configurable.pop("checkpoint_id", None)
         copy = checkpoint.copy()
         copy["channel_values"] = copy["channel_values"].copy()
         next_config = {

libs/checkpoint-postgres/langgraph/checkpoint/postgres/aio.py

@@ -240,9 +240,7 @@ class AsyncPostgresSaver(BasePostgresSaver):
         configurable = config["configurable"].copy()
         thread_id = configurable.pop("thread_id")
         checkpoint_ns = configurable.pop("checkpoint_ns")
-        checkpoint_id = configurable.pop(
-            "checkpoint_id", configurable.pop("thread_ts", None)
-        )
+        checkpoint_id = configurable.pop("checkpoint_id", None)
 
         copy = checkpoint.copy()
         copy["channel_values"] = copy["channel_values"].copy()

libs/checkpoint-postgres/langgraph/checkpoint/postgres/shallow.py

@@ -191,7 +191,7 @@ class ShallowPostgresSaver(BasePostgresSaver):
     ) -> None:
         warnings.warn(
             "ShallowPostgresSaver is deprecated as of version 2.0.20 and will be removed in 3.0.0. "
-            "Use PostgresSaver instead, and invoke the graph with `graph.invoke(..., checkpoint_during=False)`.",
+            "Use PostgresSaver instead, and invoke the graph with `graph.invoke(..., durability='exit')`.",
             DeprecationWarning,
             stacklevel=2,
         )
@@ -547,7 +547,7 @@ class AsyncShallowPostgresSaver(BasePostgresSaver):
     ) -> None:
         warnings.warn(
             "AsyncShallowPostgresSaver is deprecated as of version 2.0.20 and will be removed in 3.0.0. "
-            "Use AsyncPostgresSaver instead, and invoke the graph with `await graph.ainvoke(..., checkpoint_during=False)`.",
+            "Use AsyncPostgresSaver instead, and invoke the graph with `await graph.ainvoke(..., durability='exit')`.",
             DeprecationWarning,
             stacklevel=2,
         )

libs/checkpoint-postgres/tests/test_async.py

@@ -161,8 +161,7 @@ def test_data():
     config_1: RunnableConfig = {
         "configurable": {
             "thread_id": "thread-1",
-            # for backwards compatibility testing
-            "thread_ts": "1",
+            "checkpoint_id": "1",
             "checkpoint_ns": "",
         }
     }

libs/checkpoint-postgres/tests/test_sync.py

@@ -143,8 +143,7 @@ def test_data():
     config_1: RunnableConfig = {
         "configurable": {
             "thread_id": "thread-1",
-            # for backwards compatibility testing
-            "thread_ts": "1",
+            "checkpoint_id": "1",
             "checkpoint_ns": "",
         }
     }

libs/checkpoint-sqlite/tests/test_aiosqlite.py

@@ -19,8 +19,7 @@ class TestAsyncSqliteSaver:
         self.config_1: RunnableConfig = {
             "configurable": {
                 "thread_id": "thread-1",
-                # for backwards compatibility testing
-                "thread_ts": "1",
+                "checkpoint_id": "1",
                 "checkpoint_ns": "",
             }
         }

libs/checkpoint-sqlite/tests/test_sqlite.py

@@ -21,7 +21,7 @@ class TestSqliteSaver:
             "configurable": {
                 "thread_id": "thread-1",
                 # for backwards compatibility testing
-                "thread_ts": "1",
+                "checkpoint_id": "1",
                 "checkpoint_ns": "",
             }
         }

libs/checkpoint/README.md

@@ -36,7 +36,7 @@ Each checkpointer should conform to `langgraph.checkpoint.base.BaseCheckpointSav
 
 - `.put` - Store a checkpoint with its configuration and metadata.
 - `.put_writes` - Store intermediate writes linked to a checkpoint (i.e. pending writes).
-- `.get_tuple` - Fetch a checkpoint tuple using for a given configuration (`thread_id` and `thread_ts`).
+- `.get_tuple` - Fetch a checkpoint tuple using for a given configuration (`thread_id` and `checkpoint_id`).
 - `.list` - List checkpoints that match a given configuration and filter criteria.
 
 If the checkpointer will be used with asynchronous graph execution (i.e. executing the graph via `.ainvoke`, `.astream`, `.abatch`), checkpointer must implement asynchronous versions of the above methods (`.aput`, `.aput_writes`, `.aget_tuple`, `.alist`).
@@ -44,12 +44,12 @@ If the checkpointer will be used with asynchronous graph execution (i.e. executi
 ## Usage
 
 ```python
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 
 write_config = {"configurable": {"thread_id": "1", "checkpoint_ns": ""}}
 read_config = {"configurable": {"thread_id": "1"}}
 
-checkpointer = MemorySaver()
+checkpointer = InMemorySaver()
 checkpoint = {
     "v": 4,
     "ts": "2024-07-31T20:14:19.804150+00:00",

libs/checkpoint/langgraph/checkpoint/base/__init__.py

@@ -375,10 +375,8 @@ class EmptyChannelError(Exception):
 
 
 def get_checkpoint_id(config: RunnableConfig) -> str | None:
-    """Get checkpoint ID in a backwards-compatible manner (fallback on thread_ts)."""
-    return config["configurable"].get(
-        "checkpoint_id", config["configurable"].get("thread_ts")
-    )
+    """Get checkpoint ID."""
+    return config["configurable"].get("checkpoint_id")
 
 
 def get_checkpoint_metadata(
@@ -413,7 +411,6 @@ WRITES_IDX_MAP = {ERROR: -1, SCHEDULED: -2, INTERRUPT: -3, RESUME: -4}
 
 EXCLUDED_METADATA_KEYS = {
     "thread_id",
-    "thread_ts",
     "checkpoint_id",
     "checkpoint_ns",
     "checkpoint_map",

libs/checkpoint/tests/test_memory.py

@@ -22,8 +22,7 @@ class TestMemorySaver:
             "configurable": {
                 "thread_id": "thread-1",
                 "checkpoint_ns": "",
-                # for backwards compatibility testing
-                "thread_ts": "1",
+                "checkpoint_id": "1",
             }
         }
         self.config_2: RunnableConfig = {
@@ -190,6 +189,6 @@ class TestMemorySaver:
 
 
 def test_memory_saver() -> None:
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
-    assert isinstance(MemorySaver(), InMemorySaver)
+    assert isinstance(InMemorySaver(), InMemorySaver)

libs/cli/examples/graphs/agent.py

@@ -49,12 +49,12 @@ def call_model(state, config):
 tool_node = ToolNode(tools)
 
 
-class ConfigSchema(TypedDict):
+class ContextSchema(TypedDict):
     model: Literal["anthropic", "openai"]
 
 
 # Define a new graph
-workflow = StateGraph(AgentState, config_schema=ConfigSchema)
+workflow = StateGraph(AgentState, context_schema=ContextSchema)
 
 # Define the two nodes we will cycle between
 workflow.add_node("agent", call_model)

libs/cli/langgraph_cli/cli.py

@@ -153,6 +153,12 @@ OPT_POSTGRES_URI = click.option(
     help="Postgres URI to use for the database. Defaults to launching a local database",
 )
 
+OPT_API_VERSION = click.option(
+    "--api-version",
+    type=str,
+    help="API server version to use for the base image. If unspecified, the latest version will be used.",
+)
+
 
 @click.group()
 @click.version_option(version=__version__, prog_name="LangGraph CLI")
@@ -170,6 +176,7 @@ def cli():
 @OPT_DEBUGGER_BASE_URL
 @OPT_WATCH
 @OPT_POSTGRES_URI
+@OPT_API_VERSION
 @click.option(
     "--image",
     type=str,
@@ -203,6 +210,7 @@ def up(
     debugger_port: Optional[int],
     debugger_base_url: Optional[str],
     postgres_uri: Optional[str],
+    api_version: Optional[str],
     image: Optional[str],
     base_image: Optional[str],
 ):
@@ -225,6 +233,7 @@ For production use, requires a license key in env var LANGGRAPH_CLOUD_LICENSE_KE
             debugger_port=debugger_port,
             debugger_base_url=debugger_base_url,
             postgres_uri=postgres_uri,
+            api_version=api_version,
             image=image,
             base_image=base_image,
         )
@@ -290,6 +299,7 @@ def _build(
     config: pathlib.Path,
     config_json: dict,
     base_image: Optional[str],
+    api_version: Optional[str],
     pull: bool,
     tag: str,
     passthrough: Sequence[str] = (),
@@ -300,7 +310,7 @@ def _build(
             subp_exec(
                 "docker",
                 "pull",
-                langgraph_cli.config.docker_tag(config_json, base_image),
+                langgraph_cli.config.docker_tag(config_json, base_image, api_version),
                 verbose=True,
             )
         )
@@ -314,7 +324,7 @@ def _build(
     ]
     # apply config
     stdin, additional_contexts = langgraph_cli.config.config_to_docker(
-        config, config_json, base_image
+        config, config_json, base_image, api_version
     )
     # add additional_contexts
     if additional_contexts:
@@ -355,6 +365,7 @@ def _build(
     "\n\n    \b\nExamples:\n    --base-image langchain/langgraph-server:0.2.18  # Pin to a specific patch version"
     "\n    --base-image langchain/langgraph-server:0.2  # Pin to a minor version (Python)",
 )
+@OPT_API_VERSION
 @click.argument("docker_build_args", nargs=-1, type=click.UNPROCESSED)
 @cli.command(
     help="📦 Build LangGraph API server Docker image.",
@@ -367,6 +378,7 @@ def build(
     config: pathlib.Path,
     docker_build_args: Sequence[str],
     base_image: Optional[str],
+    api_version: Optional[str],
     pull: bool,
     tag: str,
 ):
@@ -376,7 +388,15 @@ def build(
         config_json = langgraph_cli.config.validate_config_file(config)
         warn_non_wolfi_distro(config_json)
         _build(
-            runner, set, config, config_json, base_image, pull, tag, docker_build_args
+            runner,
+            set,
+            config,
+            config_json,
+            base_image,
+            api_version,
+            pull,
+            tag,
+            docker_build_args,
         )
 
 
@@ -456,12 +476,14 @@ tests
     "\n\n    \b\nExamples:\n    --base-image langchain/langgraph-server:0.2.18  # Pin to a specific patch version"
     "\n    --base-image langchain/langgraph-server:0.2  # Pin to a minor version (Python)",
 )
+@OPT_API_VERSION
 @log_command
 def dockerfile(
     save_path: str,
     config: pathlib.Path,
     add_docker_compose: bool,
     base_image: Optional[str] = None,
+    api_version: Optional[str] = None,
 ) -> None:
     save_path = pathlib.Path(save_path).absolute()
     secho(f"🔍 Validating configuration at path: {config}", fg="yellow")
@@ -474,6 +496,7 @@ def dockerfile(
         config,
         config_json,
         base_image=base_image,
+        api_version=api_version,
     )
     with open(str(save_path), "w", encoding="utf-8") as f:
         f.write(dockerfile)
@@ -739,6 +762,7 @@ def prepare_args_and_stdin(
     debugger_port: Optional[int] = None,
     debugger_base_url: Optional[str] = None,
     postgres_uri: Optional[str] = None,
+    api_version: Optional[str] = None,
     # Like "my-tag" (if you already built it locally)
     image: Optional[str] = None,
     # Like "langchain/langgraphjs-api" or "langchain/langgraph-api
@@ -754,6 +778,7 @@ def prepare_args_and_stdin(
         postgres_uri=postgres_uri,
         image=image,  # Pass image to compose YAML generator
         base_image=base_image,
+        api_version=api_version,
     )
     args = [
         "--project-directory",
@@ -769,6 +794,7 @@ def prepare_args_and_stdin(
         config,
         watch=watch,
         base_image=langgraph_cli.config.default_base_image(config),
+        api_version=api_version,
         image=image,
     )
     return args, stdin
@@ -787,6 +813,7 @@ def prepare(
     debugger_port: Optional[int] = None,
     debugger_base_url: Optional[str] = None,
     postgres_uri: Optional[str] = None,
+    api_version: Optional[str] = None,
     image: Optional[str] = None,
     base_image: Optional[str] = None,
 ) -> tuple[list[str], str]:
@@ -799,7 +826,7 @@ def prepare(
             subp_exec(
                 "docker",
                 "pull",
-                langgraph_cli.config.docker_tag(config_json, base_image),
+                langgraph_cli.config.docker_tag(config_json, base_image, api_version),
                 verbose=verbose,
             )
         )
@@ -814,6 +841,7 @@ def prepare(
         debugger_port=debugger_port,
         debugger_base_url=debugger_base_url or f"http://127.0.0.1:{port}",
         postgres_uri=postgres_uri,
+        api_version=api_version,
         image=image,
         base_image=base_image,
     )

libs/cli/langgraph_cli/config.py

@@ -1213,6 +1213,7 @@ def python_config_to_docker(
     config_path: pathlib.Path,
     config: Config,
     base_image: str,
+    api_version: Optional[str] = None,
 ) -> tuple[str, dict[str, str]]:
     """Generate a Dockerfile from the configuration."""
     pip_installer = config.get("pip_installer", "auto")
@@ -1360,7 +1361,7 @@ ADD {relpath} /deps/{name}
                 "# -- End of JS dependencies install --",
             ]
         )
-    image_str = docker_tag(config, base_image)
+    image_str = docker_tag(config, base_image, api_version)
     docker_file_contents = [
         f"FROM {image_str}",
         "",
@@ -1402,10 +1403,11 @@ def node_config_to_docker(
     config_path: pathlib.Path,
     config: Config,
     base_image: str,
+    api_version: Optional[str] = None,
 ) -> tuple[str, dict[str, str]]:
     faux_path = f"/deps/{config_path.parent.name}"
     install_cmd = _get_node_pm_install_cmd(config_path, config)
-    image_str = docker_tag(config, base_image)
+    image_str = docker_tag(config, base_image, api_version)
 
     env_vars: list[str] = []
 
@@ -1461,6 +1463,7 @@ def default_base_image(config: Config) -> str:
 def docker_tag(
     config: Config,
     base_image: Optional[str] = None,
+    api_version: Optional[str] = None,
 ) -> str:
     base_image = base_image or default_base_image(config)
 
@@ -1473,28 +1476,43 @@ def docker_tag(
     if "/langgraph-server" in base_image:
         return f"{base_image}-py{config['python_version']}"
 
+    # Build the standard tag format
+    language, version = None, None
     if config.get("node_version") and not config.get("python_version"):
-        return f"{base_image}:{config['node_version']}{distro_tag}"
-    return f"{base_image}:{config['python_version']}{distro_tag}"
+        language, version = "node", config["node_version"]
+    else:
+        language, version = "py", config["python_version"]
+
+    version_distro_tag = f"{version}{distro_tag}"
+
+    # Prepend API version if provided
+    if api_version:
+        full_tag = f"{api_version}-{language}{version_distro_tag}"
+    else:
+        full_tag = version_distro_tag
+
+    return f"{base_image}:{full_tag}"
 
 
 def config_to_docker(
     config_path: pathlib.Path,
     config: Config,
     base_image: Optional[str] = None,
+    api_version: Optional[str] = None,
 ) -> tuple[str, dict[str, str]]:
     base_image = base_image or default_base_image(config)
 
     if config.get("node_version") and not config.get("python_version"):
-        return node_config_to_docker(config_path, config, base_image)
+        return node_config_to_docker(config_path, config, base_image, api_version)
 
-    return python_config_to_docker(config_path, config, base_image)
+    return python_config_to_docker(config_path, config, base_image, api_version)
 
 
 def config_to_compose(
     config_path: pathlib.Path,
     config: Config,
     base_image: Optional[str] = None,
+    api_version: Optional[str] = None,
     image: Optional[str] = None,
     watch: bool = False,
 ) -> str:
@@ -1531,7 +1549,7 @@ def config_to_compose(
 
     else:
         dockerfile, additional_contexts = config_to_docker(
-            config_path, config, base_image
+            config_path, config, base_image, api_version
         )
 
         additional_contexts_str = "\n".join(

libs/cli/langgraph_cli/docker.py

@@ -147,6 +147,8 @@ def compose_as_dict(
     image: Optional[str] = None,
     # Base image to use for the LangGraph API server
     base_image: Optional[str] = None,
+    # API version of the base image
+    api_version: Optional[str] = None,
 ) -> dict:
     """Create a docker compose file as a dictionary in YML style."""
     if postgres_uri is None:
@@ -252,6 +254,7 @@ def compose(
     postgres_uri: Optional[str] = None,
     image: Optional[str] = None,
     base_image: Optional[str] = None,
+    api_version: Optional[str] = None,
 ) -> str:
     """Create a docker compose file as a string."""
     compose_content = compose_as_dict(
@@ -262,6 +265,7 @@ def compose(
         postgres_uri=postgres_uri,
         image=image,
         base_image=base_image,
+        api_version=api_version,
     )
     compose_str = dict_to_yaml(compose_content)
     return compose_str

libs/cli/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "langgraph-cli"
-version = "0.3.5"
+version = "0.3.6"
 description = "CLI for interacting with LangGraph API"
 authors = []
 requires-python = ">=3.9"

libs/cli/tests/unit_tests/cli/test_cli.py

@@ -574,3 +574,248 @@ def test_build_generate_proper_build_context():
         assert len(build_contexts) == 2, (
             f"Expected 2 build contexts, but found {len(build_contexts)}"
         )
+
+
+def test_dockerfile_command_with_api_version() -> None:
+    """Test the 'dockerfile' command with --api-version flag."""
+    runner = CliRunner()
+    config_content = {
+        "python_version": "3.11",
+        "graphs": {"agent": "agent.py:graph"},
+        "dependencies": ["."],
+    }
+
+    with temporary_config_folder(config_content) as temp_dir:
+        save_path = temp_dir / "Dockerfile"
+        agent_path = temp_dir / "agent.py"
+        agent_path.touch()
+
+        result = runner.invoke(
+            cli,
+            [
+                "dockerfile",
+                str(save_path),
+                "--config",
+                str(temp_dir / "config.json"),
+                "--api-version",
+                "0.2.74",
+            ],
+        )
+
+        # Assert command was successful
+        assert result.exit_code == 0, result.output
+        assert "✅ Created: Dockerfile" in result.output
+
+        # Check if Dockerfile was created and contains correct FROM line
+        assert save_path.exists()
+        with open(save_path) as f:
+            dockerfile = f.read()
+            assert "FROM langchain/langgraph-api:0.2.74-py3.11" in dockerfile
+
+
+def test_dockerfile_command_with_api_version_and_base_image() -> None:
+    """Test the 'dockerfile' command with both --api-version and --base-image flags."""
+    runner = CliRunner()
+    config_content = {
+        "python_version": "3.12",
+        "graphs": {"agent": "agent.py:graph"},
+        "dependencies": ["."],
+        "image_distro": "wolfi",
+    }
+
+    with temporary_config_folder(config_content) as temp_dir:
+        save_path = temp_dir / "Dockerfile"
+        agent_path = temp_dir / "agent.py"
+        agent_path.touch()
+
+        result = runner.invoke(
+            cli,
+            [
+                "dockerfile",
+                str(save_path),
+                "--config",
+                str(temp_dir / "config.json"),
+                "--api-version",
+                "1.0.0",
+                "--base-image",
+                "my-registry/custom-api",
+            ],
+        )
+
+        # Assert command was successful
+        assert result.exit_code == 0, result.output
+        assert "✅ Created: Dockerfile" in result.output
+
+        # Check if Dockerfile was created and contains correct FROM line
+        assert save_path.exists()
+        with open(save_path) as f:
+            dockerfile = f.read()
+            assert "FROM my-registry/custom-api:1.0.0-py3.12-wolfi" in dockerfile
+
+
+def test_dockerfile_command_with_api_version_nodejs() -> None:
+    """Test the 'dockerfile' command with --api-version flag for Node.js config."""
+    runner = CliRunner()
+    config_content = {
+        "node_version": "20",
+        "graphs": {"agent": "agent.js:graph"},
+    }
+
+    with temporary_config_folder(config_content) as temp_dir:
+        save_path = temp_dir / "Dockerfile"
+        agent_path = temp_dir / "agent.js"
+        agent_path.touch()
+
+        result = runner.invoke(
+            cli,
+            [
+                "dockerfile",
+                str(save_path),
+                "--config",
+                str(temp_dir / "config.json"),
+                "--api-version",
+                "0.2.74",
+            ],
+        )
+
+        # Assert command was successful
+        assert result.exit_code == 0, result.output
+        assert "✅ Created: Dockerfile" in result.output
+
+        # Check if Dockerfile was created and contains correct FROM line
+        assert save_path.exists()
+        with open(save_path) as f:
+            dockerfile = f.read()
+            assert "FROM langchain/langgraphjs-api:0.2.74-node20" in dockerfile
+
+
+def test_build_command_with_api_version() -> None:
+    """Test the 'build' command with --api-version flag."""
+    runner = CliRunner()
+    config_content = {
+        "python_version": "3.11",
+        "graphs": {"agent": "agent.py:graph"},
+        "dependencies": ["."],
+        "image_distro": "wolfi",  # Use wolfi to avoid warning messages
+    }
+
+    with temporary_config_folder(config_content) as temp_dir:
+        agent_path = temp_dir / "agent.py"
+        agent_path.touch()
+
+        # Mock docker command since we don't want to actually build
+        with runner.isolated_filesystem():
+            result = runner.invoke(
+                cli,
+                [
+                    "build",
+                    "--tag",
+                    "test-image",
+                    "--config",
+                    str(temp_dir / "config.json"),
+                    "--api-version",
+                    "0.2.74",
+                    "--no-pull",  # Avoid pulling non-existent images
+                ],
+                catch_exceptions=True,
+            )
+
+        # Check that the build command is called with the correct tag
+        # The output should contain the docker build command with the api_version tag
+        assert "langchain/langgraph-api:0.2.74-py3.11-wolfi" in result.output
+
+
+def test_build_command_with_api_version_and_base_image() -> None:
+    """Test the 'build' command with both --api-version and --base-image flags."""
+    runner = CliRunner()
+    config_content = {
+        "python_version": "3.12",
+        "graphs": {"agent": "agent.py:graph"},
+        "dependencies": ["."],
+        "image_distro": "wolfi",  # Use wolfi to avoid warning messages
+    }
+
+    with temporary_config_folder(config_content) as temp_dir:
+        agent_path = temp_dir / "agent.py"
+        agent_path.touch()
+
+        # Mock docker command since we don't want to actually build
+        with runner.isolated_filesystem():
+            result = runner.invoke(
+                cli,
+                [
+                    "build",
+                    "--tag",
+                    "test-image",
+                    "--config",
+                    str(temp_dir / "config.json"),
+                    "--api-version",
+                    "1.0.0",
+                    "--base-image",
+                    "my-registry/custom-api",
+                    "--no-pull",  # Avoid pulling non-existent images
+                ],
+                catch_exceptions=True,
+            )
+
+        # Check that the build command includes the api_version
+        assert "my-registry/custom-api:1.0.0-py3.12-wolfi" in result.output
+
+
+def test_prepare_args_and_stdin_with_api_version() -> None:
+    """Test prepare_args_and_stdin function with api_version parameter."""
+    config_path = pathlib.Path(__file__).parent / "langgraph.json"
+    config = validate_config(
+        Config(dependencies=["."], graphs={"agent": "agent.py:graph"})
+    )
+    port = 8000
+    api_version = "0.2.74"
+
+    actual_args, actual_stdin = prepare_args_and_stdin(
+        capabilities=DEFAULT_DOCKER_CAPABILITIES,
+        config_path=config_path,
+        config=config,
+        docker_compose=None,
+        port=port,
+        watch=False,
+        api_version=api_version,
+    )
+
+    expected_args = [
+        "--project-directory",
+        str(pathlib.Path(__file__).parent.absolute()),
+        "-f",
+        "-",
+    ]
+
+    # Check that the args are correct
+    assert actual_args == expected_args
+
+    # Check that the stdin contains the correct FROM line with api_version
+    assert "FROM langchain/langgraph-api:0.2.74-py3.11" in actual_stdin
+
+
+def test_prepare_args_and_stdin_with_api_version_and_image() -> None:
+    """Test prepare_args_and_stdin function with both api_version and image parameters."""
+    config_path = pathlib.Path(__file__).parent / "langgraph.json"
+    config = validate_config(
+        Config(dependencies=["."], graphs={"agent": "agent.py:graph"})
+    )
+    port = 8000
+    api_version = "0.2.74"
+    image = "my-custom-image:latest"
+
+    actual_args, actual_stdin = prepare_args_and_stdin(
+        capabilities=DEFAULT_DOCKER_CAPABILITIES,
+        config_path=config_path,
+        config=config,
+        docker_compose=None,
+        port=port,
+        watch=False,
+        api_version=api_version,
+        image=image,
+    )
+
+    # When image is provided, api_version should be ignored for the image
+    # but the stdin should not contain a build section (since image is provided)
+    assert "pull_policy: build" not in actual_stdin

libs/cli/tests/unit_tests/test_config.py

@@ -1337,3 +1337,195 @@ def test_docker_tag_different_node_versions_with_distro():
         )
         tag = docker_tag(config)
         assert tag == expected_tag, f"Failed for Node.js {node_version}"
+
+
+def test_docker_tag_with_api_version():
+    """Test docker_tag function with api_version parameter."""
+
+    # Test 1: Python config with api_version and default distro
+    config = validate_config(
+        {
+            "python_version": "3.11",
+            "dependencies": ["."],
+            "graphs": {"agent": "./agent.py:graph"},
+        }
+    )
+    tag = docker_tag(config, api_version="0.2.74")
+    assert tag == "langchain/langgraph-api:0.2.74-py3.11"
+
+    # Test 2: Python config with api_version and wolfi distro
+    config = validate_config(
+        {
+            "python_version": "3.12",
+            "dependencies": ["."],
+            "graphs": {"agent": "./agent.py:graph"},
+            "image_distro": "wolfi",
+        }
+    )
+    tag = docker_tag(config, api_version="0.2.74")
+    assert tag == "langchain/langgraph-api:0.2.74-py3.12-wolfi"
+
+    # Test 3: Node.js config with api_version and default distro
+    config = validate_config(
+        {
+            "node_version": "20",
+            "graphs": {"agent": "./agent.js:graph"},
+        }
+    )
+    tag = docker_tag(config, api_version="0.2.74")
+    assert tag == "langchain/langgraphjs-api:0.2.74-node20"
+
+    # Test 4: Node.js config with api_version and wolfi distro
+    config = validate_config(
+        {
+            "node_version": "20",
+            "graphs": {"agent": "./agent.js:graph"},
+            "image_distro": "wolfi",
+        }
+    )
+    tag = docker_tag(config, api_version="0.2.74")
+    assert tag == "langchain/langgraphjs-api:0.2.74-node20-wolfi"
+
+    # Test 5: Custom base image with api_version
+    config = validate_config(
+        {
+            "python_version": "3.11",
+            "dependencies": ["."],
+            "graphs": {"agent": "./agent.py:graph"},
+            "base_image": "my-registry/custom-image",
+        }
+    )
+    tag = docker_tag(config, base_image="my-registry/custom-image", api_version="1.0.0")
+    assert tag == "my-registry/custom-image:1.0.0-py3.11"
+
+    # Test 6: api_version with different Python versions
+    for python_version in ["3.11", "3.12", "3.13"]:
+        config = validate_config(
+            {
+                "python_version": python_version,
+                "dependencies": ["."],
+                "graphs": {"agent": "./agent.py:graph"},
+            }
+        )
+        tag = docker_tag(config, api_version="0.2.74")
+        assert tag == f"langchain/langgraph-api:0.2.74-py{python_version}"
+
+    # Test 7: Without api_version should work as before
+    config = validate_config(
+        {
+            "python_version": "3.11",
+            "dependencies": ["."],
+            "graphs": {"agent": "./agent.py:graph"},
+        }
+    )
+    tag = docker_tag(config)
+    assert tag == "langchain/langgraph-api:3.11"
+
+    # Test 8: api_version with multiplatform config (should default to Python)
+    config = validate_config(
+        {
+            "python_version": "3.11",
+            "node_version": "20",
+            "dependencies": ["."],
+            "graphs": {"python": "./agent.py:graph", "js": "./agent.js:graph"},
+        }
+    )
+    tag = docker_tag(config, api_version="0.2.74")
+    assert tag == "langchain/langgraph-api:0.2.74-py3.11"
+
+    # Test 9: api_version with _INTERNAL_docker_tag should ignore api_version
+    config = validate_config(
+        {
+            "python_version": "3.11",
+            "dependencies": ["."],
+            "graphs": {"agent": "./agent.py:graph"},
+            "_INTERNAL_docker_tag": "internal-tag",
+        }
+    )
+    tag = docker_tag(config, api_version="0.2.74")
+    assert tag == "langchain/langgraph-api:internal-tag"
+
+    # Test 10: api_version with langgraph-server base image should follow special format
+    config = validate_config(
+        {
+            "python_version": "3.11",
+            "dependencies": ["."],
+            "graphs": {"agent": "./agent.py:graph"},
+        }
+    )
+    tag = docker_tag(
+        config, base_image="langchain/langgraph-server:0.2", api_version="0.2.74"
+    )
+    assert tag == "langchain/langgraph-server:0.2-py3.11"
+
+
+def test_config_to_docker_with_api_version():
+    """Test config_to_docker function with api_version parameter."""
+
+    # Test Python config with api_version
+    graphs = {"agent": "./agent.py:graph"}
+    actual_docker_stdin, additional_contexts = config_to_docker(
+        PATH_TO_CONFIG,
+        validate_config({"dependencies": ["."], "graphs": graphs}),
+        "langchain/langgraph-api",
+        api_version="0.2.74",
+    )
+
+    # Check that the FROM line uses the api_version
+    lines = actual_docker_stdin.split("\n")
+    from_line = lines[0]
+    assert from_line == "FROM langchain/langgraph-api:0.2.74-py3.11"
+
+    # Test Node.js config with api_version
+    graphs = {"agent": "./agent.js:graph"}
+    actual_docker_stdin, additional_contexts = config_to_docker(
+        PATH_TO_CONFIG,
+        validate_config({"node_version": "20", "graphs": graphs}),
+        "langchain/langgraphjs-api",
+        api_version="0.2.74",
+    )
+
+    # Check that the FROM line uses the api_version
+    lines = actual_docker_stdin.split("\n")
+    from_line = lines[0]
+    assert from_line == "FROM langchain/langgraphjs-api:0.2.74-node20"
+
+
+def test_config_to_compose_with_api_version():
+    """Test config_to_compose function with api_version parameter."""
+
+    # Test Python config with api_version
+    config = validate_config(
+        {
+            "dependencies": ["."],
+            "graphs": {"agent": "./agent.py:graph"},
+        }
+    )
+
+    actual_compose_str = config_to_compose(
+        PATH_TO_CONFIG,
+        config,
+        "langchain/langgraph-api",
+        api_version="0.2.74",
+    )
+
+    # Check that the compose file includes the correct FROM line with api_version
+    assert "FROM langchain/langgraph-api:0.2.74-py3.11" in actual_compose_str
+
+    # Test Node.js config with api_version
+    config = validate_config(
+        {
+            "node_version": "20",
+            "graphs": {"agent": "./agent.js:graph"},
+        }
+    )
+
+    actual_compose_str = config_to_compose(
+        PATH_TO_CONFIG,
+        config,
+        "langchain/langgraphjs-api",
+        api_version="0.2.74",
+    )
+
+    # Check that the compose file includes the correct FROM line with api_version
+    assert "FROM langchain/langgraphjs-api:0.2.74-node20" in actual_compose_str

libs/cli/tests/unit_tests/test_docker.py

@@ -146,3 +146,220 @@ services:
             REDIS_URI: redis://langgraph-redis:6379
             POSTGRES_URI: {DEFAULT_POSTGRES_URI}"""
     assert clean_empty_lines(actual_compose_str) == expected_compose_str
+
+
+def test_compose_with_api_version():
+    """Test compose function with api_version parameter."""
+    port = 8123
+    api_version = "0.2.74"
+
+    actual_compose_str = compose(
+        DEFAULT_DOCKER_CAPABILITIES, port=port, api_version=api_version
+    )
+
+    # The compose function should generate a compose file that doesn't directly
+    # reference the api_version, since it's handled in the docker tag creation
+    # when building the image. The compose function mainly sets up services.
+    expected_compose_str = f"""volumes:
+    langgraph-data:
+        driver: local
+services:
+    langgraph-redis:
+        image: redis:6
+        healthcheck:
+            test: redis-cli ping
+            interval: 5s
+            timeout: 1s
+            retries: 5
+    langgraph-postgres:
+        image: pgvector/pgvector:pg16
+        ports:
+            - "5433:5432"
+        environment:
+            POSTGRES_DB: postgres
+            POSTGRES_USER: postgres
+            POSTGRES_PASSWORD: postgres
+        command:
+            - postgres
+            - -c
+            - shared_preload_libraries=vector
+        volumes:
+            - langgraph-data:/var/lib/postgresql/data
+        healthcheck:
+            test: pg_isready -U postgres
+            start_period: 10s
+            timeout: 1s
+            retries: 5
+            interval: 5s
+    langgraph-api:
+        ports:
+            - "{port}:8000"
+        depends_on:
+            langgraph-redis:
+                condition: service_healthy
+            langgraph-postgres:
+                condition: service_healthy
+        environment:
+            REDIS_URI: redis://langgraph-redis:6379
+            POSTGRES_URI: {DEFAULT_POSTGRES_URI}"""
+    assert clean_empty_lines(actual_compose_str) == expected_compose_str
+
+
+def test_compose_with_api_version_and_base_image():
+    """Test compose function with both api_version and base_image parameters."""
+    port = 8123
+    api_version = "1.0.0"
+    base_image = "my-registry/custom-api"
+
+    actual_compose_str = compose(
+        DEFAULT_DOCKER_CAPABILITIES,
+        port=port,
+        api_version=api_version,
+        base_image=base_image,
+    )
+
+    # Similar to the previous test - the compose function doesn't directly embed
+    # the api_version or base_image into the compose file since those are handled
+    # during the docker build process
+    expected_compose_str = f"""volumes:
+    langgraph-data:
+        driver: local
+services:
+    langgraph-redis:
+        image: redis:6
+        healthcheck:
+            test: redis-cli ping
+            interval: 5s
+            timeout: 1s
+            retries: 5
+    langgraph-postgres:
+        image: pgvector/pgvector:pg16
+        ports:
+            - "5433:5432"
+        environment:
+            POSTGRES_DB: postgres
+            POSTGRES_USER: postgres
+            POSTGRES_PASSWORD: postgres
+        command:
+            - postgres
+            - -c
+            - shared_preload_libraries=vector
+        volumes:
+            - langgraph-data:/var/lib/postgresql/data
+        healthcheck:
+            test: pg_isready -U postgres
+            start_period: 10s
+            timeout: 1s
+            retries: 5
+            interval: 5s
+    langgraph-api:
+        ports:
+            - "{port}:8000"
+        depends_on:
+            langgraph-redis:
+                condition: service_healthy
+            langgraph-postgres:
+                condition: service_healthy
+        environment:
+            REDIS_URI: redis://langgraph-redis:6379
+            POSTGRES_URI: {DEFAULT_POSTGRES_URI}"""
+    assert clean_empty_lines(actual_compose_str) == expected_compose_str
+
+
+def test_compose_with_api_version_and_custom_postgres():
+    """Test compose function with api_version and custom postgres URI."""
+    port = 8123
+    api_version = "0.2.74"
+    custom_postgres_uri = "postgresql://user:pass@external-db:5432/mydb"
+
+    actual_compose_str = compose(
+        DEFAULT_DOCKER_CAPABILITIES,
+        port=port,
+        api_version=api_version,
+        postgres_uri=custom_postgres_uri,
+    )
+
+    expected_compose_str = f"""services:
+    langgraph-redis:
+        image: redis:6
+        healthcheck:
+            test: redis-cli ping
+            interval: 5s
+            timeout: 1s
+            retries: 5
+    langgraph-api:
+        ports:
+            - "{port}:8000"
+        depends_on:
+            langgraph-redis:
+                condition: service_healthy
+        environment:
+            REDIS_URI: redis://langgraph-redis:6379
+            POSTGRES_URI: {custom_postgres_uri}"""
+    assert clean_empty_lines(actual_compose_str) == expected_compose_str
+
+
+def test_compose_with_api_version_and_debugger():
+    """Test compose function with api_version and debugger port."""
+    port = 8123
+    debugger_port = 8001
+    api_version = "0.2.74"
+
+    actual_compose_str = compose(
+        DEFAULT_DOCKER_CAPABILITIES,
+        port=port,
+        api_version=api_version,
+        debugger_port=debugger_port,
+    )
+
+    expected_compose_str = f"""volumes:
+    langgraph-data:
+        driver: local
+services:
+    langgraph-redis:
+        image: redis:6
+        healthcheck:
+            test: redis-cli ping
+            interval: 5s
+            timeout: 1s
+            retries: 5
+    langgraph-postgres:
+        image: pgvector/pgvector:pg16
+        ports:
+            - "5433:5432"
+        environment:
+            POSTGRES_DB: postgres
+            POSTGRES_USER: postgres
+            POSTGRES_PASSWORD: postgres
+        command:
+            - postgres
+            - -c
+            - shared_preload_libraries=vector
+        volumes:
+            - langgraph-data:/var/lib/postgresql/data
+        healthcheck:
+            test: pg_isready -U postgres
+            start_period: 10s
+            timeout: 1s
+            retries: 5
+            interval: 5s
+    langgraph-debugger:
+        image: langchain/langgraph-debugger
+        restart: on-failure
+        depends_on:
+            langgraph-postgres:
+                condition: service_healthy
+        ports:
+            - "{debugger_port}:3968"
+    langgraph-api:
+        ports:
+            - "{port}:8000"
+        depends_on:
+            langgraph-redis:
+                condition: service_healthy
+            langgraph-postgres:
+                condition: service_healthy
+        environment:
+            REDIS_URI: redis://langgraph-redis:6379
+            POSTGRES_URI: {DEFAULT_POSTGRES_URI}"""
+    assert clean_empty_lines(actual_compose_str) == expected_compose_str

libs/cli/uv.lock

@@ -531,7 +531,7 @@ wheels = [
 
 [[package]]
 name = "langgraph-cli"
-version = "0.3.5"
+version = "0.3.6"
 source = { editable = "." }
 dependencies = [
     { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },

libs/langgraph/bench/__main__.py

@@ -11,7 +11,7 @@ from bench.react_agent import react_agent
 from bench.sequential import create_sequential
 from bench.wide_dict import wide_dict
 from bench.wide_state import wide_state
-from langgraph.checkpoint.memory import MemorySaver
+from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.graph import StateGraph
 from langgraph.pregel import Pregel
 
@@ -26,7 +26,7 @@ async def arun(graph: Pregel, input: dict):
                     "configurable": {"thread_id": str(uuid4())},
                     "recursion_limit": 1000000000,
                 },
-                checkpoint_during=False,
+                durability="exit",
             )
         ]
     )
@@ -43,7 +43,7 @@ async def arun_first_event_latency(graph: Pregel, input: dict) -> None:
             "configurable": {"thread_id": str(uuid4())},
             "recursion_limit": 1000000000,
         },
-        checkpoint_during=False,
+        durability="exit",
     )
 
     try:
@@ -63,7 +63,7 @@ def run(graph: Pregel, input: dict):
                     "configurable": {"thread_id": str(uuid4())},
                     "recursion_limit": 1000000000,
                 },
-                checkpoint_during=False,
+                durability="exit",
             )
         ]
     )
@@ -80,7 +80,7 @@ def run_first_event_latency(graph: Pregel, input: dict) -> None:
             "configurable": {"thread_id": str(uuid4())},
             "recursion_limit": 1000000000,
         },
-        checkpoint_during=False,
+        durability="exit",
     )
 
     try:
@@ -108,8 +108,8 @@ benchmarks = (
     ),
     (
         "fanout_to_subgraph_10x_checkpoint",
-        fanout_to_subgraph().compile(checkpointer=MemorySaver()),
-        fanout_to_subgraph_sync().compile(checkpointer=MemorySaver()),
+        fanout_to_subgraph().compile(checkpointer=InMemorySaver()),
+        fanout_to_subgraph_sync().compile(checkpointer=InMemorySaver()),
         {
             "subjects": [
                 random.choices("abcdefghijklmnopqrstuvwxyz", k=1000) for _ in range(10)
@@ -128,8 +128,8 @@ benchmarks = (
     ),
     (
         "fanout_to_subgraph_100x_checkpoint",
-        fanout_to_subgraph().compile(checkpointer=MemorySaver()),
-        fanout_to_subgraph_sync().compile(checkpointer=MemorySaver()),
+        fanout_to_subgraph().compile(checkpointer=InMemorySaver()),
+        fanout_to_subgraph_sync().compile(checkpointer=InMemorySaver()),
         {
             "subjects": [
                 random.choices("abcdefghijklmnopqrstuvwxyz", k=1000) for _ in range(100)
@@ -144,8 +144,8 @@ benchmarks = (
     ),
     (
         "react_agent_10x_checkpoint",
-        react_agent(10, checkpointer=MemorySaver()),
-        react_agent(10, checkpointer=MemorySaver()),
+        react_agent(10, checkpointer=InMemorySaver()),
+        react_agent(10, checkpointer=InMemorySaver()),
         {"messages": [HumanMessage("hi?")]},
     ),
     (
@@ -156,8 +156,8 @@ benchmarks = (
     ),
     (
         "react_agent_100x_checkpoint",
-        react_agent(100, checkpointer=MemorySaver()),
-        react_agent(100, checkpointer=MemorySaver()),
+        react_agent(100, checkpointer=InMemorySaver()),
+        react_agent(100, checkpointer=InMemorySaver()),
         {"messages": [HumanMessage("hi?")]},
     ),
     (
@@ -178,8 +178,8 @@ benchmarks = (
     ),
     (
         "wide_state_25x300_checkpoint",
-        wide_state(300).compile(checkpointer=MemorySaver()),
-        wide_state(300).compile(checkpointer=MemorySaver()),
+        wide_state(300).compile(checkpointer=InMemorySaver()),
+        wide_state(300).compile(checkpointer=InMemorySaver()),
         {
             "messages": [
                 {
@@ -210,8 +210,8 @@ benchmarks = (
     ),
     (
         "wide_state_15x600_checkpoint",
-        wide_state(600).compile(checkpointer=MemorySaver()),
-        wide_state(600).compile(checkpointer=MemorySaver()),
+        wide_state(600).compile(checkpointer=InMemorySaver()),
+        wide_state(600).compile(checkpointer=InMemorySaver()),
         {
             "messages": [
                 {
@@ -242,8 +242,8 @@ benchmarks = (
     ),
     (
         "wide_state_9x1200_checkpoint",
-        wide_state(1200).compile(checkpointer=MemorySaver()),
-        wide_state(1200).compile(checkpointer=MemorySaver()),
+        wide_state(1200).compile(checkpointer=InMemorySaver()),
+        wide_state(1200).compile(checkpointer=InMemorySaver()),
         {
             "messages": [
                 {
@@ -274,8 +274,8 @@ benchmarks = (
     ),
     (
         "wide_dict_25x300_checkpoint",
-        wide_dict(300).compile(checkpointer=MemorySaver()),
-        wide_dict(300).compile(checkpointer=MemorySaver()),
+        wide_dict(300).compile(checkpointer=InMemorySaver()),
+        wide_dict(300).compile(checkpointer=InMemorySaver()),
         {
             "messages": [
                 {
@@ -306,8 +306,8 @@ benchmarks = (
     ),
     (
         "wide_dict_15x600_checkpoint",
-        wide_dict(600).compile(checkpointer=MemorySaver()),
-        wide_dict(600).compile(checkpointer=MemorySaver()),
+        wide_dict(600).compile(checkpointer=InMemorySaver()),
+        wide_dict(600).compile(checkpointer=InMemorySaver()),
         {
             "messages": [
                 {
@@ -338,8 +338,8 @@ benchmarks = (
     ),
     (
         "wide_dict_9x1200_checkpoint",
-        wide_dict(1200).compile(checkpointer=MemorySaver()),
-        wide_dict(1200).compile(checkpointer=MemorySaver()),
+        wide_dict(1200).compile(checkpointer=InMemorySaver()),
+        wide_dict(1200).compile(checkpointer=InMemorySaver()),
         {
             "messages": [
                 {
@@ -382,8 +382,8 @@ benchmarks = (
     ),
     (
         "pydantic_state_25x300_checkpoint",
-        pydantic_state(300).compile(checkpointer=MemorySaver()),
-        pydantic_state(300).compile(checkpointer=MemorySaver()),
+        pydantic_state(300).compile(checkpointer=InMemorySaver()),
+        pydantic_state(300).compile(checkpointer=InMemorySaver()),
         {
             "messages": [
                 {
@@ -414,8 +414,8 @@ benchmarks = (
     ),
     (
         "pydantic_state_15x600_checkpoint",
-        pydantic_state(600).compile(checkpointer=MemorySaver()),
-        pydantic_state(600).compile(checkpointer=MemorySaver()),
+        pydantic_state(600).compile(checkpointer=InMemorySaver()),
+        pydantic_state(600).compile(checkpointer=InMemorySaver()),
         {
             "messages": [
                 {
@@ -446,8 +446,8 @@ benchmarks = (
     ),
     (
         "pydantic_state_9x1200_checkpoint",
-        pydantic_state(1200).compile(checkpointer=MemorySaver()),
-        pydantic_state(1200).compile(checkpointer=MemorySaver()),
+        pydantic_state(1200).compile(checkpointer=InMemorySaver()),
+        pydantic_state(1200).compile(checkpointer=InMemorySaver()),
         {
             "messages": [
                 {

libs/langgraph/bench/fanout_to_subgraph.py

@@ -3,8 +3,9 @@ from typing import Annotated
 
 from typing_extensions import TypedDict
 
-from langgraph.constants import END, START, Send
+from langgraph.constants import END, START
 from langgraph.graph.state import StateGraph
+from langgraph.types import Send
 
 
 def fanout_to_subgraph() -> StateGraph:
@@ -114,9 +115,9 @@ if __name__ == "__main__":
 
     import uvloop
 
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
-    graph = fanout_to_subgraph().compile(checkpointer=MemorySaver())
+    graph = fanout_to_subgraph().compile(checkpointer=InMemorySaver())
     input = {
         "subjects": [
             random.choices("abcdefghijklmnopqrstuvwxyz", k=1000) for _ in range(1000)

libs/langgraph/bench/pydantic_state.py

@@ -304,9 +304,9 @@ if __name__ == "__main__":
 
     import uvloop
 
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
-    graph = pydantic_state(1000).compile(checkpointer=MemorySaver())
+    graph = pydantic_state(1000).compile(checkpointer=InMemorySaver())
     input = {
         "messages": [
             {

libs/langgraph/bench/react_agent.py

@@ -68,9 +68,9 @@ if __name__ == "__main__":
 
     import uvloop
 
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
-    graph = react_agent(100, checkpointer=MemorySaver())
+    graph = react_agent(100, checkpointer=InMemorySaver())
     input = {"messages": [HumanMessage("hi?")]}
     config = {"configurable": {"thread_id": "1"}, "recursion_limit": 20000000000}
 

libs/langgraph/bench/sequential.py

@@ -1,7 +1,7 @@
 """Create a sequential no-op graph consisting of a few hundred nodes."""
 
+from langgraph._internal._runnable import RunnableCallable
 from langgraph.graph import MessagesState, StateGraph
-from langgraph.utils.runnable import RunnableCallable
 
 
 def create_sequential(number_nodes: int) -> StateGraph:

libs/langgraph/bench/wide_dict.py

@@ -130,9 +130,9 @@ if __name__ == "__main__":
 
     import uvloop
 
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
-    graph = wide_dict(1000).compile(checkpointer=MemorySaver())
+    graph = wide_dict(1000).compile(checkpointer=InMemorySaver())
     input = {
         "messages": [
             {

libs/langgraph/bench/wide_state.py

@@ -140,9 +140,9 @@ if __name__ == "__main__":
 
     import uvloop
 
-    from langgraph.checkpoint.memory import MemorySaver
+    from langgraph.checkpoint.memory import InMemorySaver
 
-    graph = wide_state(1000).compile(checkpointer=MemorySaver())
+    graph = wide_state(1000).compile(checkpointer=InMemorySaver())
     input = {
         "messages": [
             {

libs/langgraph/langgraph/_internal/__init__.py

@@ -0,0 +1,4 @@
+"""Internal modules for LangGraph.
+
+This module is not part of the public API, and thus stability is not guaranteed.
+"""

libs/langgraph/langgraph/_internal/_config.py

@@ -18,9 +18,7 @@ from langchain_core.runnables.config import (
     var_child_runnable_config,
 )
 
-from langgraph.checkpoint.base import CheckpointMetadata
-from langgraph.config import get_config, get_store, get_stream_writer  # noqa
-from langgraph.constants import (
+from langgraph._internal._constants import (
     CONF,
     CONFIG_KEY_CHECKPOINT_ID,
     CONFIG_KEY_CHECKPOINT_MAP,
@@ -28,6 +26,7 @@ from langgraph.constants import (
     NS_END,
     NS_SEP,
 )
+from langgraph.checkpoint.base import CheckpointMetadata
 
 DEFAULT_RECURSION_LIMIT = int(getenv("LANGGRAPH_DEFAULT_RECURSION_LIMIT", "25"))
 

libs/langgraph/langgraph/_internal/_constants.py

@@ -0,0 +1,110 @@
+"""Constants used for Pregel operations."""
+
+import sys
+from typing import Literal, cast
+
+# --- Reserved write keys ---
+INPUT = sys.intern("__input__")
+# for values passed as input to the graph
+INTERRUPT = sys.intern("__interrupt__")
+# for dynamic interrupts raised by nodes
+RESUME = sys.intern("__resume__")
+# for values passed to resume a node after an interrupt
+ERROR = sys.intern("__error__")
+# for errors raised by nodes
+NO_WRITES = sys.intern("__no_writes__")
+# marker to signal node didn't write anything
+TASKS = sys.intern("__pregel_tasks")
+# for Send objects returned by nodes/edges, corresponds to PUSH below
+RETURN = sys.intern("__return__")
+# for writes of a task where we simply record the return value
+PREVIOUS = sys.intern("__previous__")
+# the implicit branch that handles each node's Control values
+
+
+# --- Reserved cache namespaces ---
+CACHE_NS_WRITES = sys.intern("__pregel_ns_writes")
+# cache namespace for node writes
+
+# --- Reserved config.configurable keys ---
+CONFIG_KEY_SEND = sys.intern("__pregel_send")
+# holds the `write` function that accepts writes to state/edges/reserved keys
+CONFIG_KEY_READ = sys.intern("__pregel_read")
+# holds the `read` function that returns a copy of the current state
+CONFIG_KEY_CALL = sys.intern("__pregel_call")
+# holds the `call` function that accepts a node/func, args and returns a future
+CONFIG_KEY_CHECKPOINTER = sys.intern("__pregel_checkpointer")
+# holds a `BaseCheckpointSaver` passed from parent graph to child graphs
+CONFIG_KEY_STREAM = sys.intern("__pregel_stream")
+# holds a `StreamProtocol` passed from parent graph to child graphs
+CONFIG_KEY_CACHE = sys.intern("__pregel_cache")
+# holds a `BaseCache` made available to subgraphs
+CONFIG_KEY_RESUMING = sys.intern("__pregel_resuming")
+# holds a boolean indicating if subgraphs should resume from a previous checkpoint
+CONFIG_KEY_TASK_ID = sys.intern("__pregel_task_id")
+# holds the task ID for the current task
+CONFIG_KEY_THREAD_ID = sys.intern("thread_id")
+# holds the thread ID for the current invocation
+CONFIG_KEY_CHECKPOINT_MAP = sys.intern("checkpoint_map")
+# holds a mapping of checkpoint_ns -> checkpoint_id for parent graphs
+CONFIG_KEY_CHECKPOINT_ID = sys.intern("checkpoint_id")
+# holds the current checkpoint_id, if any
+CONFIG_KEY_CHECKPOINT_NS = sys.intern("checkpoint_ns")
+# holds the current checkpoint_ns, "" for root graph
+CONFIG_KEY_NODE_FINISHED = sys.intern("__pregel_node_finished")
+# holds a callback to be called when a node is finished
+CONFIG_KEY_SCRATCHPAD = sys.intern("__pregel_scratchpad")
+# holds a mutable dict for temporary storage scoped to the current task
+CONFIG_KEY_RUNNER_SUBMIT = sys.intern("__pregel_runner_submit")
+# holds a function that receives tasks from runner, executes them and returns results
+CONFIG_KEY_DURABILITY = sys.intern("__pregel_durability")
+# holds the durability mode, one of "sync", "async", or "exit"
+CONFIG_KEY_RUNTIME = sys.intern("__pregel_runtime")
+# holds a `Runtime` instance with context, store, stream writer, etc.
+CONFIG_KEY_RESUME_MAP = sys.intern("__pregel_resume_map")
+# holds a mapping of task ns -> resume value for resuming tasks
+
+# --- Other constants ---
+PUSH = sys.intern("__pregel_push")
+# denotes push-style tasks, ie. those created by Send objects
+PULL = sys.intern("__pregel_pull")
+# denotes pull-style tasks, ie. those triggered by edges
+NS_SEP = sys.intern("|")
+# for checkpoint_ns, separates each level (ie. graph|subgraph|subsubgraph)
+NS_END = sys.intern(":")
+# for checkpoint_ns, for each level, separates the namespace from the task_id
+CONF = cast(Literal["configurable"], sys.intern("configurable"))
+# key for the configurable dict in RunnableConfig
+NULL_TASK_ID = sys.intern("00000000-0000-0000-0000-000000000000")
+# the task_id to use for writes that are not associated with a task
+
+# redefined to avoid circular import with langgraph.constants
+_TAG_HIDDEN = sys.intern("langsmith:hidden")
+
+RESERVED = {
+    _TAG_HIDDEN,
+    # reserved write keys
+    INPUT,
+    INTERRUPT,
+    RESUME,
+    ERROR,
+    NO_WRITES,
+    # reserved config.configurable keys
+    CONFIG_KEY_SEND,
+    CONFIG_KEY_READ,
+    CONFIG_KEY_CHECKPOINTER,
+    CONFIG_KEY_STREAM,
+    CONFIG_KEY_CHECKPOINT_MAP,
+    CONFIG_KEY_RESUMING,
+    CONFIG_KEY_TASK_ID,
+    CONFIG_KEY_CHECKPOINT_MAP,
+    CONFIG_KEY_CHECKPOINT_ID,
+    CONFIG_KEY_CHECKPOINT_NS,
+    CONFIG_KEY_RESUME_MAP,
+    # other constants
+    PUSH,
+    PULL,
+    NS_SEP,
+    NS_END,
+    CONF,
+}

libs/langgraph/langgraph/_internal/_fields.py

@@ -9,9 +9,7 @@ from typing import Annotated, Any, Optional, Union, get_type_hints
 from pydantic import BaseModel
 from typing_extensions import NotRequired, ReadOnly, Required, get_origin
 
-# NOTE: this is redefined here separately from langgraph.constants
-# to avoid a circular import
-MISSING = object()
+from langgraph._internal._typing import MISSING
 
 
 def _is_optional_type(type_: Any) -> bool:

libs/langgraph/langgraph/_internal/_queue.py

@@ -128,6 +128,3 @@ class SyncQueue:
         return len(self._queue)
 
     __class_getitem__ = classmethod(types.GenericAlias)
-
-
-__all__ = ["AsyncQueue", "SyncQueue"]

libs/langgraph/langgraph/_internal/_retry.py

@@ -0,0 +1,29 @@
+def default_retry_on(exc: Exception) -> bool:
+    import httpx
+    import requests
+
+    if isinstance(exc, ConnectionError):
+        return True
+    if isinstance(exc, httpx.HTTPStatusError):
+        return 500 <= exc.response.status_code < 600
+    if isinstance(exc, requests.HTTPError):
+        return 500 <= exc.response.status_code < 600 if exc.response else True
+    if isinstance(
+        exc,
+        (
+            ValueError,
+            TypeError,
+            ArithmeticError,
+            ImportError,
+            LookupError,
+            NameError,
+            SyntaxError,
+            RuntimeError,
+            ReferenceError,
+            StopIteration,
+            StopAsyncIteration,
+            OSError,
+        ),
+    ):
+        return False
+    return True

libs/langgraph/langgraph/_internal/_runnable.py

@@ -42,20 +42,19 @@ from langchain_core.runnables.utils import Input, Output
 from langchain_core.tracers.langchain import LangChainTracer
 from typing_extensions import TypeGuard
 
-from langgraph.constants import (
-    CONF,
-    CONFIG_KEY_PREVIOUS,
-    CONFIG_KEY_STORE,
-    CONFIG_KEY_STREAM_WRITER,
-)
-from langgraph.store.base import BaseStore
-from langgraph.types import StreamWriter
-from langgraph.utils.config import (
+from langgraph._internal._config import (
     ensure_config,
     get_async_callback_manager_for_config,
     get_callback_manager_for_config,
     patch_config,
 )
+from langgraph._internal._constants import (
+    CONF,
+    CONFIG_KEY_RUNTIME,
+)
+from langgraph._internal._typing import MISSING
+from langgraph.store.base import BaseStore
+from langgraph.types import StreamWriter
 
 try:
     from langchain_core.tracers._streaming import _StreamingCallbackHandler
@@ -128,45 +127,52 @@ ANY_TYPE = object()
 
 ASYNCIO_ACCEPTS_CONTEXT = sys.version_info >= (3, 11)
 
-# List of keyword arguments that can be injected at runtime from the config object.
+# List of keyword arguments that can be injected into nodes / tasks / tools at runtime.
 # A named argument may appear multiple times if it appears with distinct types.
 KWARGS_CONFIG_KEYS: tuple[tuple[str, tuple[Any, ...], str, Any], ...] = (
     (
-        sys.intern("writer"),
+        "config",
+        (RunnableConfig, "RunnableConfig", inspect.Parameter.empty),
+        # for now, use config directly, eventually, will pop off of Runtime
+        "N/A",
+        inspect.Parameter.empty,
+    ),
+    (
+        "writer",
         (StreamWriter, "StreamWriter", inspect.Parameter.empty),
-        CONFIG_KEY_STREAM_WRITER,
+        "stream_writer",
         lambda _: None,
     ),
     (
-        # Covers store that is not optional (will raise an error if a store
-        # cannot be injected).
-        sys.intern("store"),
+        "store",
         (
             BaseStore,
             "BaseStore",
             inspect.Parameter.empty,
         ),
-        CONFIG_KEY_STORE,
+        "store",
         inspect.Parameter.empty,
     ),
     (
-        # Covers store that is optional. Will set to None if not found in config.
-        sys.intern("store"),
+        "store",
         (
             Optional[BaseStore],
-            # Best effort to catch some forward references.
-            # This will not work for cases like `"Union[None, BaseStore]"`,
-            # we'll need to re-write logic to use get_type_hints()
-            # to resolve forward references.
             "Optional[BaseStore]",
         ),
-        CONFIG_KEY_STORE,
+        "store",
         None,
     ),
     (
-        sys.intern("previous"),
+        "previous",
         (ANY_TYPE,),
-        CONFIG_KEY_PREVIOUS,
+        "previous",
+        inspect.Parameter.empty,
+    ),
+    (
+        "runtime",
+        (ANY_TYPE,),
+        # we never hit this block, we just inject runtime directly
+        "N/A",
         inspect.Parameter.empty,
     ),
 )
@@ -174,7 +180,7 @@ KWARGS_CONFIG_KEYS: tuple[tuple[str, tuple[Any, ...], str, Any], ...] = (
 config keys, default values and type annotations.
 
 Used to configure keyword arguments that can be injected at runtime
-from the config object as kwargs to `invoke`, `ainvoke`, `stream` and `astream`.
+from the `Runtime` object as kwargs to `invoke`, `ainvoke`, `stream` and `astream`.
 
 For a keyword to be injected from the config object, the function signature
 must contain a kwarg with the same name and a matching type annotation.
@@ -182,8 +188,10 @@ must contain a kwarg with the same name and a matching type annotation.
 Each tuple contains:
 - the name of the kwarg in the function signature
 - the type annotation(s) for the kwarg
-- the config key to look for the value in
-- the default value for the kwarg
+- the `Runtime` attribute for fetching the value (N/A if not applicable)
+
+This is fully internal and should be further refactored to use `get_type_hints`
+to resolve forward references and optional types formatted like BaseStore | None.
 """
 
 VALID_KINDS = (inspect.Parameter.POSITIONAL_OR_KEYWORD, inspect.Parameter.KEYWORD_ONLY)
@@ -250,7 +258,6 @@ class RunnableCallable(Runnable):
         trace: bool = True,
         recurse: bool = True,
         explode_args: bool = False,
-        func_accepts_config: bool | None = None,
         **kwargs: Any,
     ) -> None:
         self.name = name
@@ -277,31 +284,23 @@ class RunnableCallable(Runnable):
         if func is None and afunc is None:
             raise ValueError("At least one of func or afunc must be provided.")
 
-        if func_accepts_config is not None:
-            self.func_accepts_config = func_accepts_config
-            self.func_accepts: dict[str, tuple[str, Any]] = {}
-        else:
-            params = inspect.signature(cast(Callable, func or afunc)).parameters
+        self.func_accepts: dict[str, tuple[str, Any]] = {}
+        params = inspect.signature(cast(Callable, func or afunc)).parameters
 
-            self.func_accepts_config = "config" in params
-            # Mapping from kwarg name to (config key, default value) to be used.
-            # The default value is used if the config key is not found in the config.
-            self.func_accepts = {}
+        for kw, typ, runtime_key, default in KWARGS_CONFIG_KEYS:
+            p = params.get(kw)
 
-            for kw, typ, config_key, default in KWARGS_CONFIG_KEYS:
-                p = params.get(kw)
+            if p is None or p.kind not in VALID_KINDS:
+                # If parameter is not found or is not a valid kind, skip
+                continue
 
-                if p is None or p.kind not in VALID_KINDS:
-                    # If parameter is not found or is not a valid kind, skip
-                    continue
+            if typ != (ANY_TYPE,) and p.annotation not in typ:
+                # A specific type is required, but the function annotation does
+                # not match the expected type.
+                continue
 
-                if typ != (ANY_TYPE,) and p.annotation not in typ:
-                    # A specific type is required, but the function annotation does
-                    # not match the expected type.
-                    continue
-
-                # If the kwarg is accepted by the function, store the default value
-                self.func_accepts[kw] = (config_key, default)
+            # If the kwarg is accepted by the function, store the key / runtime attribute to inject
+            self.func_accepts[kw] = (runtime_key, default)
 
     def __repr__(self) -> str:
         repr_args = {
@@ -328,25 +327,33 @@ class RunnableCallable(Runnable):
         else:
             args = (input,)
             kwargs = {**self.kwargs, **kwargs}
-        if self.func_accepts_config:
-            kwargs["config"] = config
-        _conf = config[CONF]
 
-        for kw, (config_key, default_value) in self.func_accepts.items():
+        runtime = config[CONF].get(CONFIG_KEY_RUNTIME)
+
+        for kw, (runtime_key, default) in self.func_accepts.items():
             # If the kwarg is already set, use the set value
             if kw in kwargs:
                 continue
 
-            if (
-                # If the kwarg is requested, but isn't in the config AND has no
-                # default value, raise an error
-                config_key not in _conf and default_value is inspect.Parameter.empty
-            ):
-                raise ValueError(
-                    f"Missing required config key '{config_key}' for '{self.name}'."
-                )
+            kw_value: Any = MISSING
+            if kw == "config":
+                kw_value = config
+            elif runtime:
+                if kw == "runtime":
+                    kw_value = runtime
+                else:
+                    try:
+                        kw_value = getattr(runtime, runtime_key)
+                    except AttributeError:
+                        pass
 
-            kwargs[kw] = _conf.get(config_key, default_value)
+            if kw_value is MISSING:
+                if default is inspect.Parameter.empty:
+                    raise ValueError(
+                        f"Missing required config key '{runtime_key}' for '{self.name}'."
+                    )
+                kw_value = default
+            kwargs[kw] = kw_value
 
         if self.trace:
             callback_manager = get_callback_manager_for_config(config, self.tags)
@@ -392,23 +399,33 @@ class RunnableCallable(Runnable):
         else:
             args = (input,)
             kwargs = {**self.kwargs, **kwargs}
-        if self.func_accepts_config:
-            kwargs["config"] = config
-        _conf = config[CONF]
-        for kw, (config_key, default_value) in self.func_accepts.items():
+
+        runtime = config[CONF].get(CONFIG_KEY_RUNTIME)
+
+        for kw, (runtime_key, default) in self.func_accepts.items():
             # If the kwarg has already been set, use the set value
             if kw in kwargs:
                 continue
 
-            if (
-                # If the kwarg is requested, but isn't in the config AND has no
-                # default value, raise an error
-                config_key not in _conf and default_value is inspect.Parameter.empty
-            ):
-                raise ValueError(
-                    f"Missing required config key '{config_key}' for '{self.name}'."
-                )
-            kwargs[kw] = _conf.get(config_key, default_value)
+            kw_value: Any = MISSING
+            if kw == "config":
+                kw_value = config
+            elif runtime:
+                if kw == "runtime":
+                    kw_value = runtime
+                else:
+                    try:
+                        kw_value = getattr(runtime, runtime_key)
+                    except AttributeError:
+                        pass
+            if kw_value is MISSING:
+                if default is inspect.Parameter.empty:
+                    raise ValueError(
+                        f"Missing required config key '{runtime_key}' for '{self.name}'."
+                    )
+                kw_value = default
+            kwargs[kw] = kw_value
+
         if self.trace:
             callback_manager = get_async_callback_manager_for_config(config, self.tags)
             run_manager = await callback_manager.on_chain_start(

libs/langgraph/langgraph/_internal/_typing.py

@@ -42,13 +42,13 @@ It can either be a `TypedDict`, `dataclass`, or Pydantic `BaseModel`.
 Note: we cannot use either `TypedDict` or `dataclass` directly due to limitations in type checking.
 """
 
-
-class Unset:
-    """A sentinel value to represent an unset type."""
-
-
-UNSET: Unset = Unset()
+MISSING = object()
+"""Unset sentinel value."""
 
 
 class DeprecatedKwargs(TypedDict):
     """TypedDict to use for extra keyword arguments, enabling type checking warnings for deprecated arguments."""
+
+
+EMPTY_SEQ: tuple[str, ...] = tuple()
+"""An empty sequence of strings."""

libs/langgraph/langgraph/channels/__init__.py

@@ -1,15 +1,27 @@
 from langgraph.channels.any_value import AnyValue
+from langgraph.channels.base import BaseChannel
 from langgraph.channels.binop import BinaryOperatorAggregate
 from langgraph.channels.ephemeral_value import EphemeralValue
-from langgraph.channels.last_value import LastValue
+from langgraph.channels.last_value import LastValue, LastValueAfterFinish
+from langgraph.channels.named_barrier_value import (
+    NamedBarrierValue,
+    NamedBarrierValueAfterFinish,
+)
 from langgraph.channels.topic import Topic
 from langgraph.channels.untracked_value import UntrackedValue
 
-__all__ = [
+__all__ = (
+    # base
+    "BaseChannel",
+    # value types
+    "AnyValue",
     "LastValue",
-    "Topic",
-    "BinaryOperatorAggregate",
+    "LastValueAfterFinish",
     "UntrackedValue",
     "EphemeralValue",
-    "AnyValue",
-]
+    "BinaryOperatorAggregate",
+    "NamedBarrierValue",
+    "NamedBarrierValueAfterFinish",
+    # topics
+    "Topic",
+)

libs/langgraph/langgraph/channels/any_value.py

@@ -1,12 +1,16 @@
+from __future__ import annotations
+
 from collections.abc import Sequence
 from typing import Any, Generic
 
 from typing_extensions import Self
 
+from langgraph._internal._typing import MISSING
 from langgraph.channels.base import BaseChannel, Value
-from langgraph.constants import MISSING
 from langgraph.errors import EmptyChannelError
 
+__all__ = ("AnyValue",)
+
 
 class AnyValue(Generic[Value], BaseChannel[Value, Value, Value]):
     """Stores the last value received, assumes that if multiple values are
@@ -14,6 +18,8 @@ class AnyValue(Generic[Value], BaseChannel[Value, Value, Value]):
 
     __slots__ = ("typ", "value")
 
+    value: Value | Any
+
     def __init__(self, typ: Any, key: str = "") -> None:
         super().__init__(typ, key)
         self.value = MISSING

libs/langgraph/langgraph/channels/base.py

@@ -1,18 +1,22 @@
+from __future__ import annotations
+
 from abc import ABC, abstractmethod
 from collections.abc import Sequence
 from typing import Any, Generic, TypeVar
 
 from typing_extensions import Self
 
-from langgraph.constants import MISSING
-from langgraph.errors import EmptyChannelError, InvalidUpdateError
+from langgraph._internal._typing import MISSING
+from langgraph.errors import EmptyChannelError
 
 Value = TypeVar("Value")
 Update = TypeVar("Update")
-C = TypeVar("C")
+Checkpoint = TypeVar("Checkpoint")
+
+__all__ = ("BaseChannel",)
 
 
-class BaseChannel(Generic[Value, Update, C], ABC):
+class BaseChannel(Generic[Value, Update, Checkpoint], ABC):
     """Base class for all channels."""
 
     __slots__ = ("key", "typ")
@@ -39,7 +43,7 @@ class BaseChannel(Generic[Value, Update, C], ABC):
         Subclasses can override this method with a more efficient implementation."""
         return self.from_checkpoint(self.checkpoint())
 
-    def checkpoint(self) -> C:
+    def checkpoint(self) -> Checkpoint | Any:
         """Return a serializable representation of the channel's current state.
         Raises EmptyChannelError if the channel is empty (never updated yet),
         or doesn't support checkpoints."""
@@ -49,7 +53,7 @@ class BaseChannel(Generic[Value, Update, C], ABC):
             return MISSING
 
     @abstractmethod
-    def from_checkpoint(self, checkpoint: C) -> Self:
+    def from_checkpoint(self, checkpoint: Checkpoint | Any) -> Self:
         """Return a new identical channel, optionally initialized from a checkpoint.
         If the checkpoint contains complex data structures, they should be copied."""
 
@@ -99,10 +103,3 @@ class BaseChannel(Generic[Value, Update, C], ABC):
         Returns True if the channel was updated, False otherwise.
         """
         return False
-
-
-__all__ = [
-    "BaseChannel",
-    "EmptyChannelError",
-    "InvalidUpdateError",
-]

libs/langgraph/langgraph/channels/binop.py

@@ -4,10 +4,12 @@ from typing import Callable, Generic
 
 from typing_extensions import NotRequired, Required, Self
 
+from langgraph._internal._typing import MISSING
 from langgraph.channels.base import BaseChannel, Value
-from langgraph.constants import MISSING
 from langgraph.errors import EmptyChannelError
 
+__all__ = ("BinaryOperatorAggregate",)
+
 
 # Adapted from typing_extensions
 def _strip_extras(t):  # type: ignore[no-untyped-def]

libs/langgraph/langgraph/channels/ephemeral_value.py

@@ -1,18 +1,25 @@
+from __future__ import annotations
+
 from collections.abc import Sequence
 from typing import Any, Generic
 
 from typing_extensions import Self
 
+from langgraph._internal._typing import MISSING
 from langgraph.channels.base import BaseChannel, Value
-from langgraph.constants import MISSING
 from langgraph.errors import EmptyChannelError, InvalidUpdateError
 
+__all__ = ("EphemeralValue",)
+
 
 class EphemeralValue(Generic[Value], BaseChannel[Value, Value, Value]):
     """Stores the value received in the step immediately preceding, clears after."""
 
     __slots__ = ("value", "guard")
 
+    value: Value | Any
+    guard: bool
+
     def __init__(self, typ: Any, guard: bool = True) -> None:
         super().__init__(typ)
         self.guard = guard

libs/langgraph/langgraph/channels/last_value.py

@@ -1,10 +1,12 @@
+from __future__ import annotations
+
 from collections.abc import Sequence
 from typing import Any, Generic
 
 from typing_extensions import Self
 
+from langgraph._internal._typing import MISSING
 from langgraph.channels.base import BaseChannel, Value
-from langgraph.constants import MISSING
 from langgraph.errors import (
     EmptyChannelError,
     ErrorCode,
@@ -12,12 +14,16 @@ from langgraph.errors import (
     create_error_message,
 )
 
+__all__ = ("LastValue", "LastValueAfterFinish")
+
 
 class LastValue(Generic[Value], BaseChannel[Value, Value, Value]):
     """Stores the last value received, can receive at most one value per step."""
 
     __slots__ = ("value",)
 
+    value: Value | Any
+
     def __init__(self, typ: Any, key: str = "") -> None:
         super().__init__(typ, key)
         self.value = MISSING
@@ -80,6 +86,9 @@ class LastValueAfterFinish(
 
     __slots__ = ("value", "finished")
 
+    value: Value | Any
+    finished: bool
+
     def __init__(self, typ: Any, key: str = "") -> None:
         super().__init__(typ, key)
         self.value = MISSING
@@ -98,19 +107,19 @@ class LastValueAfterFinish(
         """The type of the update received by the channel."""
         return self.typ
 
-    def checkpoint(self) -> tuple[Value, bool]:
+    def checkpoint(self) -> tuple[Value | Any, bool] | Any:
         if self.value is MISSING:
             return MISSING
         return (self.value, self.finished)
 
-    def from_checkpoint(self, checkpoint: tuple[Value, bool]) -> Self:
+    def from_checkpoint(self, checkpoint: tuple[Value | Any, bool] | Any) -> Self:
         empty = self.__class__(self.typ)
         empty.key = self.key
         if checkpoint is not MISSING:
             empty.value, empty.finished = checkpoint
         return empty
 
-    def update(self, values: Sequence[Value]) -> bool:
+    def update(self, values: Sequence[Value | Any]) -> bool:
         if len(values) == 0:
             return False
 

libs/langgraph/langgraph/channels/named_barrier_value.py

@@ -3,10 +3,12 @@ from typing import Generic
 
 from typing_extensions import Self
 
+from langgraph._internal._typing import MISSING
 from langgraph.channels.base import BaseChannel, Value
-from langgraph.constants import MISSING
 from langgraph.errors import EmptyChannelError, InvalidUpdateError
 
+__all__ = ("NamedBarrierValue", "NamedBarrierValueAfterFinish")
+
 
 class NamedBarrierValue(Generic[Value], BaseChannel[Value, Value, set[Value]]):
     """A channel that waits until all named values are received before making the value available."""

libs/langgraph/langgraph/channels/topic.py

@@ -5,12 +5,14 @@ from typing import Any, Generic, Union
 
 from typing_extensions import Self
 
+from langgraph._internal._typing import MISSING
 from langgraph.channels.base import BaseChannel, Value
-from langgraph.constants import MISSING
 from langgraph.errors import EmptyChannelError
 
+__all__ = ("Topic",)
 
-def flatten(values: Sequence[Value | list[Value]]) -> Iterator[Value]:
+
+def _flatten(values: Sequence[Value | list[Value]]) -> Iterator[Value]:
     for value in values:
         if isinstance(value, list):
             yield from value
@@ -77,7 +79,7 @@ class Topic(
         if not self.accumulate:
             updated = bool(self.values)
             self.values = list[Value]()
-        if flat_values := tuple(flatten(values)):
+        if flat_values := tuple(_flatten(values)):
             updated = True
             self.values.extend(flat_values)
         return updated

libs/langgraph/langgraph/channels/untracked_value.py

@@ -1,18 +1,25 @@
+from __future__ import annotations
+
 from collections.abc import Sequence
-from typing import Generic
+from typing import Any, Generic
 
 from typing_extensions import Self
 
+from langgraph._internal._typing import MISSING
 from langgraph.channels.base import BaseChannel, Value
-from langgraph.constants import MISSING
 from langgraph.errors import EmptyChannelError, InvalidUpdateError
 
+__all__ = ("UntrackedValue",)
+
 
 class UntrackedValue(Generic[Value], BaseChannel[Value, Value, Value]):
     """Stores the last value received, never checkpointed."""
 
     __slots__ = ("value", "guard")
 
+    guard: bool
+    value: Value | Any
+
     def __init__(self, typ: type[Value], guard: bool = True) -> None:
         super().__init__(typ)
         self.guard = guard
@@ -38,7 +45,7 @@ class UntrackedValue(Generic[Value], BaseChannel[Value, Value, Value]):
         empty.value = self.value
         return empty
 
-    def checkpoint(self) -> Value:
+    def checkpoint(self) -> Value | Any:
         return MISSING
 
     def from_checkpoint(self, checkpoint: Value) -> Self:

libs/langgraph/langgraph/config.py

@@ -5,7 +5,7 @@ from typing import Any
 from langchain_core.runnables import RunnableConfig
 from langchain_core.runnables.config import var_child_runnable_config
 
-from langgraph.constants import CONF, CONFIG_KEY_STORE, CONFIG_KEY_STREAM_WRITER
+from langgraph._internal._constants import CONF, CONFIG_KEY_RUNTIME
 from langgraph.store.base import BaseStore
 from langgraph.types import StreamWriter
 
@@ -114,8 +114,7 @@ def get_store() -> BaseStore:
         3
         ```
     """
-    config = get_config()
-    return config[CONF][CONFIG_KEY_STORE]
+    return get_config()[CONF][CONFIG_KEY_RUNTIME].store
 
 
 def get_stream_writer() -> StreamWriter:
@@ -181,5 +180,5 @@ def get_stream_writer() -> StreamWriter:
         {'custom_data': 'Hello!'}
         ```
     """
-    config = get_config()
-    return config[CONF].get(CONFIG_KEY_STREAM_WRITER, _no_op_stream_writer)
+    runtime = get_config()[CONF][CONFIG_KEY_RUNTIME]
+    return runtime.stream_writer

libs/langgraph/langgraph/constants.py

@@ -1,133 +1,64 @@
 import sys
-from collections.abc import Mapping
-from types import MappingProxyType
-from typing import Any, Literal, cast
+from typing import Any
+from warnings import warn
 
-from langgraph.types import Interrupt, Send  # noqa: F401
+from langgraph._internal._constants import (
+    CONF,
+    CONFIG_KEY_CHECKPOINTER,
+    TASKS,
+)
+from langgraph.warnings import LangGraphDeprecatedSinceV10
 
-# Interrupt, Send re-exported for backwards compatibility
-
-
-# --- Empty read-only containers ---
-EMPTY_MAP: Mapping[str, Any] = MappingProxyType({})
-EMPTY_SEQ: tuple[str, ...] = tuple()
-MISSING = object()
+__all__ = (
+    "TAG_NOSTREAM",
+    "TAG_HIDDEN",
+    "START",
+    "END",
+    # retained for backwards compatibility (mostly langgraph-api), should be removed in v2 (or earlier)
+    "CONF",
+    "TASKS",
+    "CONFIG_KEY_CHECKPOINTER",
+)
 
 # --- Public constants ---
 TAG_NOSTREAM = sys.intern("nostream")
 """Tag to disable streaming for a chat model."""
-TAG_NOSTREAM_ALT = sys.intern("langsmith:nostream")
-"""Tag to disable streaming for a chat model. (Deprecated in favour of "nostream")"""
 TAG_HIDDEN = sys.intern("langsmith:hidden")
 """Tag to hide a node/edge from certain tracing/streaming environments."""
-START = sys.intern("__start__")
-"""The first (maybe virtual) node in graph-style Pregel."""
 END = sys.intern("__end__")
 """The last (maybe virtual) node in graph-style Pregel."""
-SELF = sys.intern("__self__")
-"""The implicit branch that handles each node's Control values."""
-PREVIOUS = sys.intern("__previous__")
+START = sys.intern("__start__")
+"""The first (maybe virtual) node in graph-style Pregel."""
 
-# --- Reserved write keys ---
-INPUT = sys.intern("__input__")
-# for values passed as input to the graph
-INTERRUPT = sys.intern("__interrupt__")
-# for dynamic interrupts raised by nodes
-RESUME = sys.intern("__resume__")
-# for values passed to resume a node after an interrupt
-ERROR = sys.intern("__error__")
-# for errors raised by nodes
-NO_WRITES = sys.intern("__no_writes__")
-# marker to signal node didn't write anything
-TASKS = sys.intern("__pregel_tasks")
-# for Send objects returned by nodes/edges, corresponds to PUSH below
-RETURN = sys.intern("__return__")
-# for writes of a task where we simply record the return value
 
-# --- Reserved cache namespaces ---
-CACHE_NS_WRITES = sys.intern("__pregel_ns_writes")
-# cache namespace for node writes
+def __getattr__(name: str) -> Any:
+    if name in ["Send", "Interrupt"]:
+        warn(
+            f"Importing {name} from langgraph.constants is deprecated. "
+            f"Please use 'from langgraph.types import {name}' instead.",
+            LangGraphDeprecatedSinceV10,
+            stacklevel=2,
+        )
 
-# --- Reserved config.configurable keys ---
-CONFIG_KEY_SEND = sys.intern("__pregel_send")
-# holds the `write` function that accepts writes to state/edges/reserved keys
-CONFIG_KEY_READ = sys.intern("__pregel_read")
-# holds the `read` function that returns a copy of the current state
-CONFIG_KEY_CALL = sys.intern("__pregel_call")
-# holds the `call` function that accepts a node/func, args and returns a future
-CONFIG_KEY_CHECKPOINTER = sys.intern("__pregel_checkpointer")
-# holds a `BaseCheckpointSaver` passed from parent graph to child graphs
-CONFIG_KEY_STREAM = sys.intern("__pregel_stream")
-# holds a `StreamProtocol` passed from parent graph to child graphs
-CONFIG_KEY_STREAM_WRITER = sys.intern("__pregel_stream_writer")
-# holds a `StreamWriter` for stream_mode=custom
-CONFIG_KEY_STORE = sys.intern("__pregel_store")
-# holds a `BaseStore` made available to managed values
-CONFIG_KEY_CACHE = sys.intern("__pregel_cache")
-# holds a `BaseCache` made available to subgraphs
-CONFIG_KEY_RESUMING = sys.intern("__pregel_resuming")
-# holds a boolean indicating if subgraphs should resume from a previous checkpoint
-CONFIG_KEY_TASK_ID = sys.intern("__pregel_task_id")
-# holds the task ID for the current task
-CONFIG_KEY_THREAD_ID = sys.intern("thread_id")
-# holds the thread ID for the current invocation
-CONFIG_KEY_CHECKPOINT_MAP = sys.intern("checkpoint_map")
-# holds a mapping of checkpoint_ns -> checkpoint_id for parent graphs
-CONFIG_KEY_CHECKPOINT_ID = sys.intern("checkpoint_id")
-# holds the current checkpoint_id, if any
-CONFIG_KEY_CHECKPOINT_NS = sys.intern("checkpoint_ns")
-# holds the current checkpoint_ns, "" for root graph
-CONFIG_KEY_NODE_FINISHED = sys.intern("__pregel_node_finished")
-# holds a callback to be called when a node is finished
-CONFIG_KEY_SCRATCHPAD = sys.intern("__pregel_scratchpad")
-# holds a mutable dict for temporary storage scoped to the current task
-CONFIG_KEY_PREVIOUS = sys.intern("__pregel_previous")
-# holds the previous return value from a stateful Pregel graph.
-CONFIG_KEY_RUNNER_SUBMIT = sys.intern("__pregel_runner_submit")
-# holds a function that receives tasks from runner, executes them and returns results
-CONFIG_KEY_CHECKPOINT_DURING = sys.intern("__pregel_checkpoint_during")
-# holds a boolean indicating whether to checkpoint during the run (or only at the end)
+        from importlib import import_module
 
-# --- Other constants ---
-PUSH = sys.intern("__pregel_push")
-# denotes push-style tasks, ie. those created by Send objects
-PULL = sys.intern("__pregel_pull")
-# denotes pull-style tasks, ie. those triggered by edges
-NS_SEP = sys.intern("|")
-# for checkpoint_ns, separates each level (ie. graph|subgraph|subsubgraph)
-NS_END = sys.intern(":")
-# for checkpoint_ns, for each level, separates the namespace from the task_id
-CONF = cast(Literal["configurable"], sys.intern("configurable"))
-# key for the configurable dict in RunnableConfig
-NULL_TASK_ID = sys.intern("00000000-0000-0000-0000-000000000000")
-# the task_id to use for writes that are not associated with a task
-CONFIG_KEY_RESUME_MAP = sys.intern("__pregel_resume_map")
-# holds a mapping of task ns -> resume value for resuming tasks
+        module = import_module("langgraph.types")
+        return getattr(module, name)
 
-RESERVED = {
-    TAG_HIDDEN,
-    # reserved write keys
-    INPUT,
-    INTERRUPT,
-    RESUME,
-    ERROR,
-    NO_WRITES,
-    # reserved config.configurable keys
-    CONFIG_KEY_SEND,
-    CONFIG_KEY_READ,
-    CONFIG_KEY_CHECKPOINTER,
-    CONFIG_KEY_STREAM,
-    CONFIG_KEY_STREAM_WRITER,
-    CONFIG_KEY_STORE,
-    CONFIG_KEY_RESUMING,
-    CONFIG_KEY_TASK_ID,
-    CONFIG_KEY_CHECKPOINT_MAP,
-    CONFIG_KEY_CHECKPOINT_ID,
-    CONFIG_KEY_CHECKPOINT_NS,
-    # other constants
-    PUSH,
-    PULL,
-    NS_SEP,
-    NS_END,
-    CONF,
-}
+    try:
+        from importlib import import_module
+
+        private_constants = import_module("langgraph._internal._constants")
+        attr = getattr(private_constants, name)
+        warn(
+            f"Importing {name} from langgraph.constants is deprecated. "
+            f"This constant is now private and should not be used directly. "
+            "Please let the LangGraph team know if you need this value.",
+            LangGraphDeprecatedSinceV10,
+            stacklevel=2,
+        )
+        return attr
+    except AttributeError:
+        pass
+
+    raise AttributeError(f"module has no attribute '{name}'")

libs/langgraph/langgraph/errors.py

@@ -1,11 +1,29 @@
+from __future__ import annotations
+
 from collections.abc import Sequence
 from enum import Enum
 from typing import Any
+from warnings import warn
 
+from typing_extensions import deprecated
+
+# EmptyChannelError is re-exported from langgraph.channels.base
 from langgraph.checkpoint.base import EmptyChannelError  # noqa: F401
 from langgraph.types import Command, Interrupt
+from langgraph.warnings import LangGraphDeprecatedSinceV10
 
-# EmptyChannelError re-exported for backwards compatibility
+__all__ = (
+    "EmptyChannelError",
+    "ErrorCode",
+    "GraphRecursionError",
+    "InvalidUpdateError",
+    "GraphBubbleUp",
+    "GraphInterrupt",
+    "NodeInterrupt",
+    "ParentCommand",
+    "EmptyInputError",
+    "TaskNotFound",
+)
 
 
 class ErrorCode(Enum):
@@ -71,11 +89,26 @@ class GraphInterrupt(GraphBubbleUp):
         super().__init__(interrupts)
 
 
+@deprecated(
+    "NodeInterrupt is deprecated. Please use `langgraph.types.interrupt` instead.",
+    stacklevel=2,
+)
 class NodeInterrupt(GraphInterrupt):
-    """Raised by a node to interrupt execution."""
+    """Raised by a node to interrupt execution.
 
-    def __init__(self, value: Any) -> None:
-        super().__init__([Interrupt(value=value)])
+    Deprecated in V1.0.0 in favor of [`interrupt`][langgraph.types.interrupt].
+    """
+
+    def __init__(self, value: Any, id: str | None = None) -> None:
+        warn(
+            "NodeInterrupt is deprecated. Please use `langgraph.types.interrupt` instead.",
+            LangGraphDeprecatedSinceV10,
+            stacklevel=2,
+        )
+        if id is None:
+            super().__init__([Interrupt(value=value)])
+        else:
+            super().__init__([Interrupt(value=value, id=id)])
 
 
 class ParentCommand(GraphBubbleUp):

libs/langgraph/langgraph/func/__init__.py

@@ -12,6 +12,7 @@ from typing import (
     Callable,
     Generic,
     TypeVar,
+    cast,
     get_args,
     get_origin,
     overload,
@@ -19,14 +20,15 @@ from typing import (
 
 from typing_extensions import Unpack
 
-from langgraph._typing import UNSET, DeprecatedKwargs
+from langgraph._internal._constants import CACHE_NS_WRITES, PREVIOUS
+from langgraph._internal._typing import MISSING, DeprecatedKwargs
 from langgraph.cache.base import BaseCache
 from langgraph.channels.ephemeral_value import EphemeralValue
 from langgraph.channels.last_value import LastValue
 from langgraph.checkpoint.base import BaseCheckpointSaver
-from langgraph.constants import CACHE_NS_WRITES, END, PREVIOUS, START
+from langgraph.constants import END, START
 from langgraph.pregel import Pregel
-from langgraph.pregel.call import (
+from langgraph.pregel._call import (
     P,
     SyncAsyncFuture,
     T,
@@ -34,14 +36,17 @@ from langgraph.pregel.call import (
     get_runnable_for_entrypoint,
     identifier,
 )
-from langgraph.pregel.read import PregelNode
-from langgraph.pregel.write import ChannelWrite, ChannelWriteEntry
+from langgraph.pregel._read import PregelNode
+from langgraph.pregel._write import ChannelWrite, ChannelWriteEntry
 from langgraph.store.base import BaseStore
 from langgraph.types import _DC_KWARGS, CachePolicy, RetryPolicy, StreamMode
-from langgraph.warnings import LangGraphDeprecatedSinceV05
+from langgraph.typing import ContextT
+from langgraph.warnings import LangGraphDeprecatedSinceV05, LangGraphDeprecatedSinceV10
+
+__all__ = ("task", "entrypoint")
 
 
-class TaskFunction(Generic[P, T]):
+class _TaskFunction(Generic[P, T]):
     def __init__(
         self,
         func: Callable[P, T],
@@ -97,14 +102,14 @@ def task(
     **kwargs: Unpack[DeprecatedKwargs],
 ) -> Callable[
     [Callable[P, Awaitable[T]] | Callable[P, T]],
-    TaskFunction[P, T],
+    _TaskFunction[P, T],
 ]: ...
 
 
 @overload
 def task(
     __func_or_none__: Callable[P, Awaitable[T]] | Callable[P, T],
-) -> TaskFunction[P, T]: ...
+) -> _TaskFunction[P, T]: ...
 
 
 def task(
@@ -115,8 +120,8 @@ def task(
     cache_policy: CachePolicy[Callable[P, str | bytes]] | None = None,
     **kwargs: Unpack[DeprecatedKwargs],
 ) -> (
-    Callable[[Callable[P, Awaitable[T]] | Callable[P, T]], TaskFunction[P, T]]
-    | TaskFunction[P, T]
+    Callable[[Callable[P, Awaitable[T]] | Callable[P, T]], _TaskFunction[P, T]]
+    | _TaskFunction[P, T]
 ):
     """Define a LangGraph task using the `task` decorator.
 
@@ -176,7 +181,7 @@ def task(
         await add_one.ainvoke([1, 2, 3])  # Returns [2, 3, 4]
         ```
     """
-    if (retry := kwargs.get("retry", UNSET)) is not UNSET:
+    if (retry := kwargs.get("retry", MISSING)) is not MISSING:
         warnings.warn(
             "`retry` is deprecated and will be removed. Please use `retry_policy` instead.",
             category=LangGraphDeprecatedSinceV05,
@@ -196,7 +201,7 @@ def task(
     def decorator(
         func: Callable[P, Awaitable[T]] | Callable[P, T],
     ) -> Callable[P, concurrent.futures.Future[T]] | Callable[P, asyncio.Future[T]]:
-        return TaskFunction(
+        return _TaskFunction(
             func, retry_policy=retry_policies, cache_policy=cache_policy, name=name
         )
 
@@ -214,7 +219,7 @@ S = TypeVar("S")
 # In this form, the `final` attribute should play nicely with IDE autocompletion,
 # and type checking tools.
 # In addition, we'll be able to surface this information in the API Reference.
-class entrypoint:
+class entrypoint(Generic[ContextT]):
     """Define a LangGraph workflow using the `entrypoint` decorator.
 
     ### Function signature
@@ -230,10 +235,9 @@ class entrypoint:
 
     | Parameter        | Description                                                                                        |
     |------------------|----------------------------------------------------------------------------------------------------|
-    | **`store`**      | An instance of [BaseStore][langgraph.store.base.BaseStore]. Useful for long-term memory.           |
-    | **`writer`**     | A [StreamWriter][langgraph.types.StreamWriter] instance for writing custom data to a stream.       |
     | **`config`**     | A configuration object (aka RunnableConfig) that holds run-time configuration values.              |
     | **`previous`**   | The previous return value for the given thread (available only when a checkpointer is provided).   |
+    | **`runtime`**    | A Runtime object that contains information about the current run, including context, store, writer |                                |
 
     The entrypoint decorator can be applied to sync functions or async functions.
 
@@ -253,7 +257,7 @@ class entrypoint:
         store: A generalized key-value store. Some implementations may support
             semantic search capabilities through an optional `index` configuration.
         cache: A cache to use for caching the results of the workflow.
-        config_schema: Specifies the schema for the configuration object that will be
+        context_schema: Specifies the schema for the context object that will be
             passed to the workflow.
         cache_policy: A cache policy to use for caching the results of the workflow.
         retry_policy: A retry policy (or list of policies) to use for the workflow in case of a failure.
@@ -264,14 +268,14 @@ class entrypoint:
 
         from langgraph.func import entrypoint, task
         from langgraph.types import interrupt, Command
-        from langgraph.checkpoint.memory import MemorySaver
+        from langgraph.checkpoint.memory import InMemorySaver
 
         @task
         def compose_essay(topic: str) -> str:
             time.sleep(1.0)  # Simulate slow operation
             return f"An essay about {topic}"
 
-        @entrypoint(checkpointer=MemorySaver())
+        @entrypoint(checkpointer=InMemorySaver())
         def review_workflow(topic: str) -> dict:
             \"\"\"Manages the workflow for generating and reviewing an essay.
 
@@ -326,10 +330,10 @@ class entrypoint:
         of the previous invocation on the same thread id.
 
         ```python
-        from langgraph.checkpoint.memory import MemorySaver
+        from langgraph.checkpoint.memory import InMemorySaver
         from langgraph.func import entrypoint
 
-        @entrypoint(checkpointer=MemorySaver())
+        @entrypoint(checkpointer=InMemorySaver())
         def my_workflow(input_data: str, previous: Optional[str] = None) -> str:
             return "world"
 
@@ -348,10 +352,10 @@ class entrypoint:
         long as the same thread id is used.
 
         ```python
-        from langgraph.checkpoint.memory import MemorySaver
+        from langgraph.checkpoint.memory import InMemorySaver
         from langgraph.func import entrypoint
 
-        @entrypoint(checkpointer=MemorySaver())
+        @entrypoint(checkpointer=InMemorySaver())
         def my_workflow(number: int, *, previous: Any = None) -> entrypoint.final[int, int]:
             previous = previous or 0
             # This will return the previous value to the caller, saving
@@ -375,27 +379,36 @@ class entrypoint:
         checkpointer: BaseCheckpointSaver | None = None,
         store: BaseStore | None = None,
         cache: BaseCache | None = None,
-        config_schema: type[Any] | None = None,
+        context_schema: type[ContextT] | None = None,
         cache_policy: CachePolicy | None = None,
         retry_policy: RetryPolicy | Sequence[RetryPolicy] | None = None,
         **kwargs: Unpack[DeprecatedKwargs],
     ) -> None:
         """Initialize the entrypoint decorator."""
-        if (retry := kwargs.get("retry", UNSET)) is not UNSET:
+        if (config_schema := kwargs.get("config_schema", MISSING)) is not MISSING:
+            warnings.warn(
+                "`config_schema` is deprecated and will be removed. Please use `context_schema` instead.",
+                category=LangGraphDeprecatedSinceV10,
+                stacklevel=2,
+            )
+            if context_schema is None:
+                context_schema = cast(type[ContextT], config_schema)
+
+        if (retry := kwargs.get("retry", MISSING)) is not MISSING:
             warnings.warn(
                 "`retry` is deprecated and will be removed. Please use `retry_policy` instead.",
                 category=LangGraphDeprecatedSinceV05,
                 stacklevel=2,
             )
             if retry_policy is None:
-                retry_policy = retry  # type: ignore[assignment]
+                retry_policy = cast("RetryPolicy | Sequence[RetryPolicy]", retry)
 
         self.checkpointer = checkpointer
         self.store = store
         self.cache = cache
         self.cache_policy = cache_policy
         self.retry_policy = retry_policy
-        self.config_schema = config_schema
+        self.context_schema = context_schema
 
     @dataclass(**_DC_KWARGS)
     class final(Generic[R, S]):
@@ -406,10 +419,10 @@ class entrypoint:
 
         Example: Decoupling the return value and the save value
             ```python
-            from langgraph.checkpoint.memory import MemorySaver
+            from langgraph.checkpoint.memory import InMemorySaver
             from langgraph.func import entrypoint
 
-            @entrypoint(checkpointer=MemorySaver())
+            @entrypoint(checkpointer=InMemorySaver())
             def my_workflow(number: int, *, previous: Any = None) -> entrypoint.final[int, int]:
                 previous = previous or 0
                 # This will return the previous value to the caller, saving
@@ -527,5 +540,5 @@ class entrypoint:
             cache=self.cache,
             cache_policy=self.cache_policy,
             retry_policy=self.retry_policy or (),
-            config_type=self.config_schema,
+            context_schema=self.context_schema,  # type: ignore[arg-type]
         )

libs/langgraph/langgraph/graph/__init__.py

@@ -2,11 +2,11 @@ from langgraph.constants import END, START
 from langgraph.graph.message import MessageGraph, MessagesState, add_messages
 from langgraph.graph.state import StateGraph
 
-__all__ = [
+__all__ = (
     "END",
     "START",
     "StateGraph",
-    "MessageGraph",
     "add_messages",
     "MessagesState",
-]
+    "MessageGraph",
+)

libs/langgraph/langgraph/graph/_branch.py

@@ -26,15 +26,15 @@ from langchain_core.runnables import (
     RunnableLambda,
 )
 
-from langgraph.constants import END, START
-from langgraph.errors import InvalidUpdateError
-from langgraph.pregel.write import PASSTHROUGH, ChannelWrite, ChannelWriteEntry
-from langgraph.types import Send
-from langgraph.utils.runnable import (
+from langgraph._internal._runnable import (
     RunnableCallable,
 )
+from langgraph.constants import END, START
+from langgraph.errors import InvalidUpdateError
+from langgraph.pregel._write import PASSTHROUGH, ChannelWrite, ChannelWriteEntry
+from langgraph.types import Send
 
-Writer = Callable[
+_Writer = Callable[
     [Sequence[Union[str, Send]], bool],
     Sequence[Union[ChannelWriteEntry, Send]],
 ]
@@ -82,7 +82,7 @@ def _get_branch_path_input_schema(
     return input
 
 
-class Branch(NamedTuple):
+class BranchSpec(NamedTuple):
     path: Runnable[Any, Hashable | list[Hashable]]
     ends: dict[Hashable, str] | None
     input_schema: type[Any] | None = None
@@ -93,7 +93,7 @@ class Branch(NamedTuple):
         path: Runnable[Any, Hashable | list[Hashable]],
         path_map: dict[Hashable, str] | list[str] | None,
         infer_schema: bool = False,
-    ) -> Branch:
+    ) -> BranchSpec:
         # coerce path_map to a dictionary
         path_map_: dict[Hashable, str] | None = None
         try:
@@ -123,7 +123,7 @@ class Branch(NamedTuple):
 
     def run(
         self,
-        writer: Writer,
+        writer: _Writer,
         reader: Callable[[RunnableConfig], Any] | None = None,
     ) -> RunnableCallable:
         return ChannelWrite.register_writer(
@@ -134,7 +134,6 @@ class Branch(NamedTuple):
                 reader=reader,
                 name=None,
                 trace=False,
-                func_accepts_config=True,
             ),
             list(
                 zip_longest(
@@ -152,7 +151,7 @@ class Branch(NamedTuple):
         config: RunnableConfig,
         *,
         reader: Callable[[RunnableConfig], Any] | None,
-        writer: Writer,
+        writer: _Writer,
     ) -> Runnable:
         if reader:
             value = reader(config)
@@ -175,7 +174,7 @@ class Branch(NamedTuple):
         config: RunnableConfig,
         *,
         reader: Callable[[RunnableConfig], Any] | None,
-        writer: Writer,
+        writer: _Writer,
     ) -> Runnable:
         if reader:
             value = reader(config)
@@ -194,7 +193,7 @@ class Branch(NamedTuple):
 
     def _finish(
         self,
-        writer: Writer,
+        writer: _Writer,
         input: Any,
         result: Any,
         config: RunnableConfig,

libs/langgraph/langgraph/graph/_node.py

@@ -0,0 +1,96 @@
+from __future__ import annotations
+
+import sys
+from collections.abc import Sequence
+from dataclasses import dataclass
+from typing import Any, Generic, Protocol, Union
+
+from langchain_core.runnables import Runnable, RunnableConfig
+from typing_extensions import TypeAlias
+
+from langgraph._internal._typing import EMPTY_SEQ
+from langgraph.runtime import Runtime
+from langgraph.store.base import BaseStore
+from langgraph.types import CachePolicy, RetryPolicy, StreamWriter
+from langgraph.typing import ContextT, NodeInputT, NodeInputT_contra
+
+_DC_SLOTS = {"slots": True} if sys.version_info >= (3, 10) else {}
+
+
+class _Node(Protocol[NodeInputT_contra]):
+    def __call__(self, state: NodeInputT_contra) -> Any: ...
+
+
+class _NodeWithConfig(Protocol[NodeInputT_contra]):
+    def __call__(self, state: NodeInputT_contra, config: RunnableConfig) -> Any: ...
+
+
+class _NodeWithWriter(Protocol[NodeInputT_contra]):
+    def __call__(self, state: NodeInputT_contra, *, writer: StreamWriter) -> Any: ...
+
+
+class _NodeWithStore(Protocol[NodeInputT_contra]):
+    def __call__(self, state: NodeInputT_contra, *, store: BaseStore) -> Any: ...
+
+
+class _NodeWithWriterStore(Protocol[NodeInputT_contra]):
+    def __call__(
+        self, state: NodeInputT_contra, *, writer: StreamWriter, store: BaseStore
+    ) -> Any: ...
+
+
+class _NodeWithConfigWriter(Protocol[NodeInputT_contra]):
+    def __call__(
+        self, state: NodeInputT_contra, *, config: RunnableConfig, writer: StreamWriter
+    ) -> Any: ...
+
+
+class _NodeWithConfigStore(Protocol[NodeInputT_contra]):
+    def __call__(
+        self, state: NodeInputT_contra, *, config: RunnableConfig, store: BaseStore
+    ) -> Any: ...
+
+
+class _NodeWithConfigWriterStore(Protocol[NodeInputT_contra]):
+    def __call__(
+        self,
+        state: NodeInputT_contra,
+        *,
+        config: RunnableConfig,
+        writer: StreamWriter,
+        store: BaseStore,
+    ) -> Any: ...
+
+
+class _NodeWithRuntime(Protocol[NodeInputT_contra, ContextT]):
+    def __call__(
+        self, state: NodeInputT_contra, *, runtime: Runtime[ContextT]
+    ) -> Any: ...
+
+
+# TODO: we probably don't want to explicitly support the config / store signatures once
+# we move to adding a context arg. Maybe what we do is we add support for kwargs with param spec
+# this is purely for typing purposes though, so can easily change in the coming weeks.
+StateNode: TypeAlias = Union[
+    _Node[NodeInputT],
+    _NodeWithConfig[NodeInputT],
+    _NodeWithWriter[NodeInputT],
+    _NodeWithStore[NodeInputT],
+    _NodeWithWriterStore[NodeInputT],
+    _NodeWithConfigWriter[NodeInputT],
+    _NodeWithConfigStore[NodeInputT],
+    _NodeWithConfigWriterStore[NodeInputT],
+    _NodeWithRuntime[NodeInputT, ContextT],
+    Runnable[NodeInputT, Any],
+]
+
+
+@dataclass(**_DC_SLOTS)
+class StateNodeSpec(Generic[NodeInputT, ContextT]):
+    runnable: StateNode[NodeInputT, ContextT]
+    metadata: dict[str, Any] | None
+    input_schema: type[NodeInputT]
+    retry_policy: RetryPolicy | Sequence[RetryPolicy] | None
+    cache_policy: CachePolicy | None
+    ends: tuple[str, ...] | dict[str, str] | None = EMPTY_SEQ
+    defer: bool = False

libs/langgraph/langgraph/graph/message.py

@@ -24,9 +24,15 @@ from langchain_core.messages import (
 )
 from typing_extensions import TypedDict
 
-from langgraph.constants import CONF, CONFIG_KEY_SEND
+from langgraph._internal._constants import CONF, CONFIG_KEY_SEND, NS_SEP
 from langgraph.graph.state import StateGraph
 
+__all__ = (
+    "add_messages",
+    "MessagesState",
+    "MessageGraph",
+)
+
 Messages = Union[list[MessageLikeRepresentation], MessageLikeRepresentation]
 
 REMOVE_ALL_MESSAGES = "__remove_all__"
@@ -314,8 +320,7 @@ def push_message(
     )
 
     from langgraph.config import get_config
-    from langgraph.constants import NS_SEP
-    from langgraph.pregel.messages import StreamMessagesHandler
+    from langgraph.pregel._messages import StreamMessagesHandler
 
     config = get_config()
     message = next(x for x in convert_to_messages([message]))