Update headers type

This PR updates the OpenAPI specification with changes detected from the
LangGraph API server.

**Changes detected as of LangGraph API version 0.4.9**

This update was automatically generated by the sync workflow in the
langgraph-api repository.

Co-authored-by: hinthornw <hinthornw@users.noreply.github.com>

docs/docs/cloud/deployment/cloud.md

@@ -99,8 +99,8 @@ Starting from the `LangGraph Platform` view...
 1. In the top-right corner, select the gear icon (`Deployment Settings`).
 1. Update the `Git Branch` to the desired branch.
 1. Check/uncheck checkbox to `Automatically update deployment on push to branch`.
-    1. Branch creation/deletion and tag creation/deletion events will not trigger an update. Only pushes to an existing branch will trigger an update.
-    1. Pushes in quick succession to a branch will not trigger subsequent updates. In the future, this functionality may be changed/improved.
+1. Branch creation/deletion and tag creation/deletion events will not trigger an update. Only pushes to an existing branch will trigger an update.
+1. Pushes in quick succession to a branch will queue subsequent updates. Once a build completes, the most recent commit will begin building and the other queued builds will be skipped.
 
 ## Add or Remove GitHub Repositories
 

docs/docs/cloud/reference/api/openapi.json

@@ -29,6 +29,10 @@
       "name": "Store",
       "description": "Store is an API for managing persistent key-value store (long-term memory) that is available from any thread."
     },
+    {
+      "name": "A2A",
+      "description": "Agent-to-Agent Protocol related endpoints for exposing assistants as A2A-compliant agents."
+    },
     {
       "name": "MCP",
       "description": "Model Context Protocol related endpoints for exposing an agent as an MCP server."
@@ -1520,6 +1524,96 @@
         }
       }
     },
+    "/threads/{thread_id}/stream": {
+      "get": {
+        "tags": [
+          "Threads"
+        ],
+        "summary": "Join Thread Stream",
+        "description": "This endpoint streams output in real-time from a thread. The stream will include the output of each run executed sequentially on the thread and will remain open indefinitely. It is the responsibility of the calling client to close the connection.",
+        "operationId": "join_thread_stream_threads__thread_id__stream_get",
+        "parameters": [
+          {
+            "description": "The ID of the thread.",
+            "required": true,
+            "schema": {
+              "type": "string",
+              "format": "uuid",
+              "title": "Thread Id",
+              "description": "The ID of the thread."
+            },
+            "name": "thread_id",
+            "in": "path"
+          },
+          {
+            "required": false,
+            "schema": {
+              "type": "string",
+              "title": "Last Event ID",
+              "description": "The ID of the last event received. Used to resume streaming from a specific point. Pass '-' to resume from the beginning."
+            },
+            "name": "Last-Event-ID",
+            "in": "header"
+          },
+          {
+            "required": false,
+            "schema": {
+              "anyOf": [
+                {
+                  "type": "string",
+                  "enum": ["lifecycle", "run_modes", "state_update"]
+                },
+                {
+                  "type": "array",
+                  "items": {
+                    "type": "string",
+                    "enum": ["lifecycle", "run_modes", "state_update"]
+                  }
+                }
+              ],
+              "default": ["run_modes"],
+              "title": "Stream Modes",
+              "description": "Stream modes to control which events are returned. 'lifecycle' returns only run start/end events, 'run_modes' returns all run events (default behavior), 'state_update' returns only state update events."
+            },
+            "name": "stream_modes",
+            "in": "query"
+          }
+        ],
+        "responses": {
+          "200": {
+            "description": "Success",
+            "content": {
+              "text/event-stream": {
+                "schema": {
+                  "type": "string",
+                  "description": "The server will send a stream of events in SSE format.\n\n**Example event**:\n\nid: 1\n\nevent: message\n\ndata: {}"
+                }
+              }
+            }
+          },
+          "404": {
+            "description": "Not Found",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/ErrorResponse"
+                }
+              }
+            }
+          },
+          "422": {
+            "description": "Validation Error",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "$ref": "#/components/schemas/ErrorResponse"
+                }
+              }
+            }
+          }
+        }
+      }
+    },
     "/threads/{thread_id}/runs": {
       "get": {
         "tags": [
@@ -3092,6 +3186,195 @@
         }
       }
     },
+    "/a2a/{assistant_id}": {
+      "post": {
+        "operationId": "post_a2a",
+        "summary": "A2A Post",
+        "description": "Communicate with an assistant using the Agent-to-Agent Protocol.\nSends a JSON-RPC 2.0 message to the assistant.\n\n- **Request**: Provide an object with `jsonrpc`, `id`, `method`, and optional `params`.\n- **Response**: Returns a JSON-RPC response with task information or error.\n\n**Supported Methods:**\n- `message/send`: Send a message to the assistant\n- `tasks/get`: Get the status and result of a task\n\n**Notes:**\n- Supports threaded conversations via thread context\n- Messages can contain text and data parts\n- Tasks run asynchronously and return completion status\n",
+        "parameters": [
+          {
+            "name": "assistant_id",
+            "in": "path",
+            "required": true,
+            "schema": {
+              "type": "string",
+              "format": "uuid"
+            },
+            "description": "The ID of the assistant to communicate with"
+          },
+          {
+            "name": "Accept",
+            "in": "header",
+            "required": true,
+            "schema": {
+              "type": "string",
+              "enum": ["application/json"]
+            },
+            "description": "Must be application/json"
+          }
+        ],
+        "requestBody": {
+          "required": true,
+          "content": {
+            "application/json": {
+              "schema": {
+                "type": "object",
+                "properties": {
+                  "jsonrpc": {
+                    "type": "string",
+                    "enum": ["2.0"],
+                    "description": "JSON-RPC version"
+                  },
+                  "id": {
+                    "type": "string",
+                    "description": "Request identifier"
+                  },
+                  "method": {
+                    "type": "string",
+                    "enum": ["message/send", "tasks/get"],
+                    "description": "The method to invoke"
+                  },
+                  "params": {
+                    "type": "object",
+                    "description": "Method parameters",
+                    "oneOf": [
+                      {
+                        "title": "Message Send Parameters",
+                        "properties": {
+                          "message": {
+                            "type": "object",
+                            "properties": {
+                              "role": {
+                                "type": "string",
+                                "enum": ["user", "assistant"],
+                                "description": "Message role"
+                              },
+                              "parts": {
+                                "type": "array",
+                                "items": {
+                                  "oneOf": [
+                                    {
+                                      "title": "Text Part",
+                                      "type": "object",
+                                      "properties": {
+                                        "kind": {
+                                          "type": "string",
+                                          "enum": ["text"]
+                                        },
+                                        "text": {
+                                          "type": "string"
+                                        }
+                                      },
+                                      "required": ["kind", "text"]
+                                    },
+                                    {
+                                      "title": "Data Part", 
+                                      "type": "object",
+                                      "properties": {
+                                        "kind": {
+                                          "type": "string",
+                                          "enum": ["data"]
+                                        },
+                                        "data": {
+                                          "type": "object"
+                                        }
+                                      },
+                                      "required": ["kind", "data"]
+                                    }
+                                  ]
+                                },
+                                "description": "Message parts"
+                              },
+                              "messageId": {
+                                "type": "string",
+                                "description": "Unique message identifier"
+                              }
+                            },
+                            "required": ["role", "parts", "messageId"]
+                          },
+                          "thread": {
+                            "type": "object",
+                            "properties": {
+                              "threadId": {
+                                "type": "string",
+                                "description": "Thread identifier for conversation context"
+                              }
+                            },
+                            "description": "Optional thread context"
+                          }
+                        },
+                        "required": ["message"]
+                      },
+                      {
+                        "title": "Task Get Parameters",
+                        "properties": {
+                          "taskId": {
+                            "type": "string",
+                            "description": "Task identifier to retrieve"
+                          }
+                        },
+                        "required": ["taskId"]
+                      }
+                    ]
+                  }
+                },
+                "required": ["jsonrpc", "id", "method"]
+              }
+            }
+          }
+        },
+        "responses": {
+          "200": {
+            "description": "JSON-RPC response",
+            "content": {
+              "application/json": {
+                "schema": {
+                  "type": "object",
+                  "properties": {
+                    "jsonrpc": {
+                      "type": "string",
+                      "enum": ["2.0"]
+                    },
+                    "id": {
+                      "type": "string"
+                    },
+                    "result": {
+                      "type": "object",
+                      "description": "Success result containing task information or task details"
+                    },
+                    "error": {
+                      "type": "object",
+                      "properties": {
+                        "code": {
+                          "type": "integer"
+                        },
+                        "message": {
+                          "type": "string"
+                        }
+                      },
+                      "description": "Error information if request failed"
+                    }
+                  },
+                  "required": ["jsonrpc", "id"]
+                }
+              }
+            }
+          },
+          "400": {
+            "description": "Bad request - invalid JSON-RPC or missing Accept header"
+          },
+          "404": {
+            "description": "Assistant not found"
+          },
+          "500": {
+            "description": "Internal server error"
+          }
+        },
+        "tags": [
+          "A2A"
+        ]
+      }
+    },
     "/mcp/": {
       "post": {
         "operationId": "post_mcp",
@@ -4346,6 +4629,17 @@
             "title": "Checkpoint During",
             "description": "Whether to checkpoint during the run.",
             "default": false
+          },
+          "durability": {
+            "type": "string",
+            "enum": [
+              "sync",
+              "async",
+              "exit"
+            ],
+            "title": "Durability",
+            "description": "Durability level for the run. Must be one of 'sync', 'async', or 'exit'.",
+            "default": "async"
           }
         },
         "type": "object",
@@ -4582,6 +4876,17 @@
             "title": "Checkpoint During",
             "description": "Whether to checkpoint during the run.",
             "default": false
+          },
+          "durability": {
+            "type": "string",
+            "enum": [
+              "sync",
+              "async",
+              "exit"
+            ],
+            "title": "Durability",
+            "description": "Durability level for the run. Must be one of 'sync', 'async', or 'exit'.",
+            "default": "async"
           }
         },
         "type": "object",
@@ -4710,6 +5015,12 @@
       },
       "ThreadSearchRequest": {
         "properties": {
+          "ids": {
+            "type": "array",
+            "items": {"type": "string", "format": "uuid"},
+            "title": "Ids",
+            "description": "List of thread IDs to include. Others are excluded."
+          },
           "metadata": {
             "type": "object",
             "title": "Metadata",

docs/docs/concepts/auth.md

@@ -112,7 +112,7 @@ auth = Auth()
 @auth.authenticate
 async def authenticate(headers: dict) -> Auth.types.MinimalUserDict:
     # Validate credentials (e.g., API key, JWT token)
-    api_key = headers.get("x-api-key")
+    api_key = headers.get(b"x-api-key")
     if not api_key or not is_valid_key(api_key):
         raise Auth.exceptions.HTTPException(
             status_code=401,

docs/docs/concepts/low_level.md

@@ -1040,7 +1040,7 @@ def node_a(state: State, runtime: Runtime[ContextSchema]):
     ...
 ```
 
-See [this guide](../how-tos/graph-api.ipynb#add-runtime-configuration) for a full breakdown on configuration.
+See [this guide](../how-tos/graph-api.md#add-runtime-configuration) for a full breakdown on configuration.
 :::
 
 :::js

docs/docs/how-tos/auth/custom_auth.md

@@ -43,7 +43,7 @@ To leverage custom authentication and access user-level metadata in your deploym
 
     @auth.authenticate # (1)!
     async def authenticate(headers: dict) -> Auth.types.MinimalUserDict:
-        api_key = headers.get("x-api-key")
+        api_key = headers.get(b"x-api-key")
         if not api_key or not is_valid_key(api_key):
             raise Auth.exceptions.HTTPException(status_code=401, detail="Invalid API key")
 

libs/cli/langgraph_cli/config.py

@@ -357,6 +357,8 @@ class HttpConfig(TypedDict, total=False):
 
     You can include or exclude headers as configurable values to condition your
     agent's behavior or permissions on a request's headers."""
+    logging_headers: Optional[ConfigurableHeaderConfig]
+    """Optional. Defines which headers are excluded from logging."""
 
 
 class Config(TypedDict, total=False):

libs/cli/schemas/schema.json

@@ -576,6 +576,17 @@
         "disable_threads": {
           "type": "boolean",
           "description": "Optional. If True, /threads routes are removed.\n\nDefault is False.\n"
+        },
+        "logging_headers": {
+          "anyOf": [
+            {
+              "$ref": "#/$defs/ConfigurableHeaderConfig"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "description": "Optional. Defines which headers are excluded from logging."
         }
       },
       "required": []

libs/cli/schemas/schema.v0.json

@@ -576,6 +576,17 @@
         "disable_threads": {
           "type": "boolean",
           "description": "Optional. If True, /threads routes are removed.\n\nDefault is False.\n"
+        },
+        "logging_headers": {
+          "anyOf": [
+            {
+              "$ref": "#/$defs/ConfigurableHeaderConfig"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "description": "Optional. Defines which headers are excluded from logging."
         }
       },
       "required": []

libs/sdk-py/langgraph_sdk/__init__.py

@@ -1,6 +1,6 @@
 from langgraph_sdk.auth import Auth
 from langgraph_sdk.client import get_client, get_sync_client
 
-__version__ = "0.2.3"
+__version__ = "0.2.5"
 
 __all__ = ["Auth", "get_client", "get_sync_client"]

libs/sdk-py/langgraph_sdk/auth/types.py

@@ -489,6 +489,9 @@ class ThreadsSearch(typing.TypedDict, total=False):
     offset: int
     """Offset for pagination."""
 
+    ids: Sequence[UUID] | None
+    """typing.Optional list of thread IDs to filter by."""
+
     thread_id: UUID | None
     """typing.Optional thread ID to filter by."""
 

libs/sdk-py/langgraph_sdk/client.py

@@ -15,6 +15,7 @@ import logging
 import os
 import re
 import sys
+import warnings
 from collections.abc import AsyncIterator, Iterator, Mapping, Sequence
 from types import TracebackType
 from typing import (
@@ -45,6 +46,7 @@ from langgraph_sdk.schema import (
     CronSelectField,
     CronSortBy,
     DisconnectMode,
+    Durability,
     GraphSchema,
     IfNotExists,
     Item,
@@ -69,6 +71,7 @@ from langgraph_sdk.schema import (
     ThreadSortBy,
     ThreadState,
     ThreadStatus,
+    ThreadStreamMode,
     ThreadUpdateStateResponse,
 )
 from langgraph_sdk.sse import SSEDecoder, aiter_lines_raw, iter_lines_raw
@@ -1306,6 +1309,7 @@ class ThreadsClient:
         *,
         metadata: Json = None,
         values: Json = None,
+        ids: Sequence[str] | None = None,
         status: ThreadStatus | None = None,
         limit: int = 10,
         offset: int = 0,
@@ -1320,6 +1324,7 @@ class ThreadsClient:
         Args:
             metadata: Thread metadata to filter on.
             values: State values to filter on.
+            ids: List of thread IDs to filter by.
             status: Thread status to filter on.
                 Must be one of 'idle', 'busy', 'interrupted' or 'error'.
             limit: Limit on number of threads to return.
@@ -1353,6 +1358,8 @@ class ThreadsClient:
             payload["metadata"] = metadata
         if values:
             payload["values"] = values
+        if ids:
+            payload["ids"] = ids
         if status:
             payload["status"] = status
         if sort_by:
@@ -1682,6 +1689,53 @@ class ThreadsClient:
             params=params,
         )
 
+    async def join_stream(
+        self,
+        thread_id: str,
+        *,
+        last_event_id: str | None = None,
+        stream_mode: ThreadStreamMode | Sequence[ThreadStreamMode] = "run_modes",
+        headers: Mapping[str, str] | None = None,
+        params: QueryParamTypes | None = None,
+    ) -> AsyncIterator[StreamPart]:
+        """Get a stream of events for a thread.
+
+        Args:
+            thread_id: The ID of the thread to get the stream for.
+            last_event_id: The ID of the last event to get.
+            headers: Optional custom headers to include with the request.
+            params: Optional query parameters to include with the request.
+
+        Returns:
+            Iterator[StreamPart]: An iterator of stream parts.
+
+        ???+ example "Example Usage"
+
+            ```python
+
+            for chunk in client.threads.join_stream(
+                thread_id="my_thread_id",
+                last_event_id="my_event_id",
+            ):
+                print(chunk)
+            ```
+
+        """  # noqa: E501
+        query_params = {
+            "stream_mode": stream_mode,
+        }
+        if params:
+            query_params.update(params)
+        return self.http.stream(
+            f"/threads/{thread_id}/stream",
+            "GET",
+            headers={
+                **({"Last-Event-ID": last_event_id} if last_event_id else {}),
+                **(headers or {}),
+            },
+            params=query_params,
+        )
+
 
 class RunsClient:
     """Client for managing runs in LangGraph.
@@ -1772,7 +1826,7 @@ class RunsClient:
         context: Context | None = None,
         checkpoint: Checkpoint | None = None,
         checkpoint_id: str | None = None,
-        checkpoint_during: bool | None = None,
+        checkpoint_during: bool | None = None,  # deprecated
         interrupt_before: All | Sequence[str] | None = None,
         interrupt_after: All | Sequence[str] | None = None,
         feedback_keys: Sequence[str] | None = None,
@@ -1785,6 +1839,7 @@ class RunsClient:
         headers: Mapping[str, str] | None = None,
         params: QueryParamTypes | None = None,
         on_run_created: Callable[[RunCreateMetadata], None] | None = None,
+        durability: Durability | None = None,
     ) -> AsyncIterator[StreamPart]:
         """Create a run and stream the results.
 
@@ -1804,7 +1859,7 @@ class RunsClient:
             context: Static context to add to the assistant.
                 !!! version-added "Supported with langgraph>=0.6.0"
             checkpoint: The checkpoint to resume from.
-            checkpoint_during: Whether to checkpoint during the run (or only at the end/interruption).
+            checkpoint_during: (deprecated) Whether to checkpoint during the run (or only at the end/interruption).
             interrupt_before: Nodes to interrupt immediately before they get executed.
             interrupt_after: Nodes to Nodes to interrupt immediately after they get executed.
             feedback_keys: Feedback keys to assign to run.
@@ -1822,6 +1877,10 @@ class RunsClient:
             headers: Optional custom headers to include with the request.
             params: Optional query parameters to include with the request.
             on_run_created: Callback when a run is created.
+            durability: The durability to use for the run. Values are "sync", "async", or "exit".
+                "async" means checkpoints are persisted async while next graph step executes, replaces checkpoint_during=True
+                "sync" means checkpoints are persisted sync after graph step executes, replaces checkpoint_during=False
+                "exit" means checkpoints are only persisted when the run exits, does not save intermediate steps
 
         Returns:
             AsyncIterator[StreamPart]: Asynchronous iterator of stream results.
@@ -1857,6 +1916,13 @@ class RunsClient:
             ```
 
         """  # noqa: E501
+        if checkpoint_during is not None:
+            warnings.warn(
+                "`checkpoint_during` is deprecated and will be removed in a future version. Use `durability` instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
+
         payload = {
             "input": input,
             "command": (
@@ -1881,6 +1947,7 @@ class RunsClient:
             "on_disconnect": on_disconnect,
             "on_completion": on_completion,
             "after_seconds": after_seconds,
+            "durability": durability,
         }
         endpoint = (
             f"/threads/{thread_id}/runs/stream"
@@ -1971,7 +2038,7 @@ class RunsClient:
         context: Context | None = None,
         checkpoint: Checkpoint | None = None,
         checkpoint_id: str | None = None,
-        checkpoint_during: bool | None = None,
+        checkpoint_during: bool | None = None,  # deprecated
         interrupt_before: All | Sequence[str] | None = None,
         interrupt_after: All | Sequence[str] | None = None,
         webhook: str | None = None,
@@ -1982,6 +2049,7 @@ class RunsClient:
         headers: Mapping[str, str] | None = None,
         params: QueryParamTypes | None = None,
         on_run_created: Callable[[RunCreateMetadata], None] | None = None,
+        durability: Durability | None = None,
     ) -> Run:
         """Create a background run.
 
@@ -2001,7 +2069,7 @@ class RunsClient:
             context: Static context to add to the assistant.
                 !!! version-added "Supported with langgraph>=0.6.0"
             checkpoint: The checkpoint to resume from.
-            checkpoint_during: Whether to checkpoint during the run (or only at the end/interruption).
+            checkpoint_during: (deprecated) Whether to checkpoint during the run (or only at the end/interruption).
             interrupt_before: Nodes to interrupt immediately before they get executed.
             interrupt_after: Nodes to Nodes to interrupt immediately after they get executed.
             webhook: Webhook to call after LangGraph API call is done.
@@ -2015,6 +2083,10 @@ class RunsClient:
                 Use to schedule future runs.
             headers: Optional custom headers to include with the request.
             on_run_created: Optional callback to call when a run is created.
+            durability: The durability to use for the run. Values are "sync", "async", or "exit".
+                "async" means checkpoints are persisted async while next graph step executes, replaces checkpoint_during=True
+                "sync" means checkpoints are persisted sync after graph step executes, replaces checkpoint_during=False
+                "exit" means checkpoints are only persisted when the run exits, does not save intermediate steps
 
         Returns:
             Run: The created background run.
@@ -2090,6 +2162,12 @@ class RunsClient:
             }
             ```
         """  # noqa: E501
+        if checkpoint_during is not None:
+            warnings.warn(
+                "`checkpoint_during` is deprecated and will be removed in a future version. Use `durability` instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
         payload = {
             "input": input,
             "command": (
@@ -2112,6 +2190,7 @@ class RunsClient:
             "if_not_exists": if_not_exists,
             "on_completion": on_completion,
             "after_seconds": after_seconds,
+            "durability": durability,
         }
         payload = {k: v for k, v in payload.items() if v is not None}
 
@@ -2209,7 +2288,7 @@ class RunsClient:
         context: Context | None = None,
         checkpoint: Checkpoint | None = None,
         checkpoint_id: str | None = None,
-        checkpoint_during: bool | None = None,
+        checkpoint_during: bool | None = None,  # deprecated
         interrupt_before: All | Sequence[str] | None = None,
         interrupt_after: All | Sequence[str] | None = None,
         webhook: str | None = None,
@@ -2222,6 +2301,7 @@ class RunsClient:
         headers: Mapping[str, str] | None = None,
         params: QueryParamTypes | None = None,
         on_run_created: Callable[[RunCreateMetadata], None] | None = None,
+        durability: Durability | None = None,
     ) -> list[dict] | dict[str, Any]:
         """Create a run, wait until it finishes and return the final state.
 
@@ -2237,7 +2317,7 @@ class RunsClient:
             context: Static context to add to the assistant.
                 !!! version-added "Supported with langgraph>=0.6.0"
             checkpoint: The checkpoint to resume from.
-            checkpoint_during: Whether to checkpoint during the run (or only at the end/interruption).
+            checkpoint_during: (deprecated) Whether to checkpoint during the run (or only at the end/interruption).
             interrupt_before: Nodes to interrupt immediately before they get executed.
             interrupt_after: Nodes to Nodes to interrupt immediately after they get executed.
             webhook: Webhook to call after LangGraph API call is done.
@@ -2253,6 +2333,10 @@ class RunsClient:
                 Use to schedule future runs.
             headers: Optional custom headers to include with the request.
             on_run_created: Optional callback to call when a run is created.
+            durability: The durability to use for the run. Values are "sync", "async", or "exit".
+                "async" means checkpoints are persisted async while next graph step executes, replaces checkpoint_during=True
+                "sync" means checkpoints are persisted sync after graph step executes, replaces checkpoint_during=False
+                "exit" means checkpoints are only persisted when the run exits, does not save intermediate steps
 
         Returns:
             Union[list[dict], dict[str, Any]]: The output of the run.
@@ -2306,6 +2390,12 @@ class RunsClient:
             ```
 
         """  # noqa: E501
+        if checkpoint_during is not None:
+            warnings.warn(
+                "`checkpoint_during` is deprecated and will be removed in a future version. Use `durability` instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
         payload = {
             "input": input,
             "command": (
@@ -2326,6 +2416,7 @@ class RunsClient:
             "on_disconnect": on_disconnect,
             "on_completion": on_completion,
             "after_seconds": after_seconds,
+            "durability": durability,
         }
         endpoint = (
             f"/threads/{thread_id}/runs/wait" if thread_id is not None else "/runs/wait"
@@ -4367,6 +4458,7 @@ class SyncThreadsClient:
         *,
         metadata: Json = None,
         values: Json = None,
+        ids: Sequence[str] | None = None,
         status: ThreadStatus | None = None,
         limit: int = 10,
         offset: int = 0,
@@ -4381,6 +4473,7 @@ class SyncThreadsClient:
         Args:
             metadata: Thread metadata to filter on.
             values: State values to filter on.
+            ids: List of thread IDs to filter by.
             status: Thread status to filter on.
                 Must be one of 'idle', 'busy', 'interrupted' or 'error'.
             limit: Limit on number of threads to return.
@@ -4410,6 +4503,8 @@ class SyncThreadsClient:
             payload["metadata"] = metadata
         if values:
             payload["values"] = values
+        if ids:
+            payload["ids"] = ids
         if status:
             payload["status"] = status
         if sort_by:
@@ -4733,6 +4828,54 @@ class SyncThreadsClient:
             params=params,
         )
 
+    def join_stream(
+        self,
+        thread_id: str,
+        *,
+        stream_mode: ThreadStreamMode | Sequence[ThreadStreamMode] = "run_modes",
+        last_event_id: str | None = None,
+        headers: Mapping[str, str] | None = None,
+        params: QueryParamTypes | None = None,
+    ) -> Iterator[StreamPart]:
+        """Get a stream of events for a thread.
+
+        Args:
+            thread_id: The ID of the thread to get the stream for.
+            last_event_id: The ID of the last event to get.
+            headers: Optional custom headers to include with the request.
+            params: Optional query parameters to include with the request.
+
+        Returns:
+            Iterator[StreamPart]: An iterator of stream parts.
+
+        ???+ example "Example Usage"
+
+            ```python
+
+            for chunk in client.threads.join_stream(
+                thread_id="my_thread_id",
+                last_event_id="my_event_id",
+                stream_mode="run_modes",
+            ):
+                print(chunk)
+            ```
+
+        """  # noqa: E501
+        query_params = {
+            "stream_mode": stream_mode,
+        }
+        if params:
+            query_params.update(params)
+        return self.http.stream(
+            f"/threads/{thread_id}/stream",
+            "GET",
+            headers={
+                **({"Last-Event-ID": last_event_id} if last_event_id else {}),
+                **(headers or {}),
+            },
+            params=query_params,
+        )
+
 
 class SyncRunsClient:
     """Synchronous client for managing runs in LangGraph.
@@ -4823,7 +4966,7 @@ class SyncRunsClient:
         context: Context | None = None,
         checkpoint: Checkpoint | None = None,
         checkpoint_id: str | None = None,
-        checkpoint_during: bool | None = None,
+        checkpoint_during: bool | None = None,  # deprecated
         interrupt_before: All | Sequence[str] | None = None,
         interrupt_after: All | Sequence[str] | None = None,
         feedback_keys: Sequence[str] | None = None,
@@ -4836,6 +4979,7 @@ class SyncRunsClient:
         headers: Mapping[str, str] | None = None,
         params: QueryParamTypes | None = None,
         on_run_created: Callable[[RunCreateMetadata], None] | None = None,
+        durability: Durability | None = None,
     ) -> Iterator[StreamPart]:
         """Create a run and stream the results.
 
@@ -4855,7 +4999,7 @@ class SyncRunsClient:
             context: Static context to add to the assistant.
                 !!! version-added "Supported with langgraph>=0.6.0"
             checkpoint: The checkpoint to resume from.
-            checkpoint_during: Whether to checkpoint during the run (or only at the end/interruption).
+            checkpoint_during: (deprecated) Whether to checkpoint during the run (or only at the end/interruption).
             interrupt_before: Nodes to interrupt immediately before they get executed.
             interrupt_after: Nodes to Nodes to interrupt immediately after they get executed.
             feedback_keys: Feedback keys to assign to run.
@@ -4872,6 +5016,11 @@ class SyncRunsClient:
                 Use to schedule future runs.
             headers: Optional custom headers to include with the request.
             on_run_created: Optional callback to call when a run is created.
+            durability: The durability to use for the run. Values are "sync", "async", or "exit".
+                "async" means checkpoints are persisted async while next graph step executes, replaces checkpoint_during=True
+                "sync" means checkpoints are persisted sync after graph step executes, replaces checkpoint_during=False
+                "exit" means checkpoints are only persisted when the run exits, does not save intermediate steps
+
 
         Returns:
             Iterator[StreamPart]: Iterator of stream results.
@@ -4904,6 +5053,12 @@ class SyncRunsClient:
             StreamPart(event='end', data=None)
             ```
         """  # noqa: E501
+        if checkpoint_during is not None:
+            warnings.warn(
+                "`checkpoint_during` is deprecated and will be removed in a future version. Use `durability` instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
         payload = {
             "input": input,
             "command": (
@@ -4928,6 +5083,7 @@ class SyncRunsClient:
             "on_disconnect": on_disconnect,
             "on_completion": on_completion,
             "after_seconds": after_seconds,
+            "durability": durability,
         }
         endpoint = (
             f"/threads/{thread_id}/runs/stream"
@@ -5018,7 +5174,7 @@ class SyncRunsClient:
         context: Context | None = None,
         checkpoint: Checkpoint | None = None,
         checkpoint_id: str | None = None,
-        checkpoint_during: bool | None = None,
+        checkpoint_during: bool | None = None,  # deprecated
         interrupt_before: All | Sequence[str] | None = None,
         interrupt_after: All | Sequence[str] | None = None,
         webhook: str | None = None,
@@ -5029,6 +5185,7 @@ class SyncRunsClient:
         headers: Mapping[str, str] | None = None,
         params: QueryParamTypes | None = None,
         on_run_created: Callable[[RunCreateMetadata], None] | None = None,
+        durability: Durability | None = None,
     ) -> Run:
         """Create a background run.
 
@@ -5048,7 +5205,7 @@ class SyncRunsClient:
             context: Static context to add to the assistant.
                 !!! version-added "Supported with langgraph>=0.6.0"
             checkpoint: The checkpoint to resume from.
-            checkpoint_during: Whether to checkpoint during the run (or only at the end/interruption).
+            checkpoint_during: (deprecated) Whether to checkpoint during the run (or only at the end/interruption).
             interrupt_before: Nodes to interrupt immediately before they get executed.
             interrupt_after: Nodes to Nodes to interrupt immediately after they get executed.
             webhook: Webhook to call after LangGraph API call is done.
@@ -5062,6 +5219,10 @@ class SyncRunsClient:
                 Use to schedule future runs.
             headers: Optional custom headers to include with the request.
             on_run_created: Optional callback to call when a run is created.
+            durability: The durability to use for the run. Values are "sync", "async", or "exit".
+                "async" means checkpoints are persisted async while next graph step executes, replaces checkpoint_during=True
+                "sync" means checkpoints are persisted sync after graph step executes, replaces checkpoint_during=False
+                "exit" means checkpoints are only persisted when the run exits, does not save intermediate steps
 
         Returns:
             Run: The created background run.
@@ -5137,6 +5298,12 @@ class SyncRunsClient:
             }
             ```
         """  # noqa: E501
+        if checkpoint_during is not None:
+            warnings.warn(
+                "`checkpoint_during` is deprecated and will be removed in a future version. Use `durability` instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
         payload = {
             "input": input,
             "command": (
@@ -5159,6 +5326,7 @@ class SyncRunsClient:
             "if_not_exists": if_not_exists,
             "on_completion": on_completion,
             "after_seconds": after_seconds,
+            "durability": durability,
         }
         payload = {k: v for k, v in payload.items() if v is not None}
 
@@ -5254,7 +5422,7 @@ class SyncRunsClient:
         metadata: Mapping[str, Any] | None = None,
         config: Config | None = None,
         context: Context | None = None,
-        checkpoint_during: bool | None = None,
+        checkpoint_during: bool | None = None,  # deprecated
         checkpoint: Checkpoint | None = None,
         checkpoint_id: str | None = None,
         interrupt_before: All | Sequence[str] | None = None,
@@ -5269,6 +5437,7 @@ class SyncRunsClient:
         headers: Mapping[str, str] | None = None,
         params: QueryParamTypes | None = None,
         on_run_created: Callable[[RunCreateMetadata], None] | None = None,
+        durability: Durability | None = None,
     ) -> list[dict] | dict[str, Any]:
         """Create a run, wait until it finishes and return the final state.
 
@@ -5284,7 +5453,7 @@ class SyncRunsClient:
             context: Static context to add to the assistant.
                 !!! version-added "Supported with langgraph>=0.6.0"
             checkpoint: The checkpoint to resume from.
-            checkpoint_during: Whether to checkpoint during the run (or only at the end/interruption).
+            checkpoint_during: (deprecated) Whether to checkpoint during the run (or only at the end/interruption).
             interrupt_before: Nodes to interrupt immediately before they get executed.
             interrupt_after: Nodes to Nodes to interrupt immediately after they get executed.
             webhook: Webhook to call after LangGraph API call is done.
@@ -5301,6 +5470,10 @@ class SyncRunsClient:
             raise_error: Whether to raise an error if the run fails.
             headers: Optional custom headers to include with the request.
             on_run_created: Optional callback to call when a run is created.
+            durability: The durability to use for the run. Values are "sync", "async", or "exit".
+                "async" means checkpoints are persisted async while next graph step executes, replaces checkpoint_during=True
+                "sync" means checkpoints are persisted sync after graph step executes, replaces checkpoint_during=False
+                "exit" means checkpoints are only persisted when the run exits, does not save intermediate steps
 
         Returns:
             Union[list[dict], dict[str, Any]]: The output of the run.
@@ -5355,6 +5528,12 @@ class SyncRunsClient:
             ```
 
         """  # noqa: E501
+        if checkpoint_during is not None:
+            warnings.warn(
+                "`checkpoint_during` is deprecated and will be removed in a future version. Use `durability` instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
         payload = {
             "input": input,
             "command": (
@@ -5376,6 +5555,7 @@ class SyncRunsClient:
             "on_completion": on_completion,
             "after_seconds": after_seconds,
             "raise_error": raise_error,
+            "durability": durability,
         }
 
         def on_response(res: httpx.Response):

libs/sdk-py/langgraph_sdk/schema.py

@@ -38,6 +38,14 @@ Represents the status of a thread:
 - "error": An exception occurred during task processing.
 """
 
+ThreadStreamMode = Literal["run_modes", "lifecycle", "state_update"]
+"""
+Defines the mode of streaming:
+- "run_modes": Stream the same events as the runs on thread, as well as run_done events.
+- "lifecycle": Stream only run start/end events.
+- "state_update": Stream state updates on the thread.
+"""
+
 StreamMode = Literal[
     "values",
     "messages",
@@ -91,6 +99,12 @@ Defines action after completion:
 - "keep": Retain resources after completion.
 """
 
+Durability = Literal["sync", "async", "exit"]
+"""Durability mode for the graph execution.
+- `"sync"`: Changes are persisted synchronously before the next step starts.
+- `"async"`: Changes are persisted asynchronously while the next step executes.
+- `"exit"`: Changes are persisted only when the graph exits."""
+
 All = Literal["*"]
 """Represents a wildcard or 'all' selector."""