WEBDIENSTE/langgraph
1
0
Fork 0
mirror of https://github.com/langchain-ai/langgraph.git synced 2026-08-21 07:02:25 +02:00
Code Issues Packages Projects Releases Wiki Activity
5,227 Commits 321 Branches 571 Tags
4aeaffef4e0d14de09fa4e34636372cb01992c6a
Commit Graph
2446 Commits
Author SHA1 Message Date
William FHandGitHub 4aeaffef4e 0.4.2 (#4570) 
Release
 2025-05-06 18:28:37 -07:00
vbarda f1bfd6051a update 2025-05-05 21:02:10 -04:00
vbarda 998be75f34 langgraph: decouple name from assistant ID in RemoteGraph 2025-05-05 20:47:22 -04:00
Nuno Campos 365dd5f459 sdk-py: Prefix private functions in sdk with _ 2025-05-05 14:43:34 -07:00
Vadym BardaandGitHub 0e61cc2cf6 prebuilt: remove state_modifier (#4439) 
Removes support for `state_modifier` param in `create_react_agent`. To
specify system prompt / instructions use `prompt` parameter.
 2025-05-05 09:49:28 -04:00
Sydney RunkleandGitHub a0a302dec5 prebuilt: switch to executing parallel tool calls via Send by default (#4438) 2025-05-05 09:49:09 -04:00
William FHandGitHub 4e05db8537 Release checkpoint-sqlite (#4509) 2025-05-01 22:42:07 -07:00
William FHandGitHub d6e20e6d09 Add missing 'running' RunStatus (#4508) 2025-05-02 05:36:58 +00:00
William FHandGitHub fcd06acd33 Add support for specifying a custom base image in docker commands (#4500) 
build & dockerfile commands

can be specified via CLI  > langgraph.json
 2025-05-01 23:13:49 +00:00
Sydney Runkle 94994a8d99 meaningless commit to get vercel going sigh 2025-05-01 13:50:55 -04:00
Sydney Runkle f35caf3401 fix for langgraph 2025-05-01 13:41:18 -04:00
Sydney Runkle 3cf2291354 docstrings for prebuilt 2025-05-01 13:40:58 -04:00
Sydney Runkle 0f96441a7c docstrings for sdk-py 2025-05-01 13:40:39 -04:00
Sydney Runkle 6c6bfbc63a docstrings for checkpoint-sqlite 2025-05-01 13:40:21 -04:00
Sydney Runkle 8ce33b948c docstrings for checkpoint-postgres 2025-05-01 13:40:05 -04:00
Sydney Runkle cc698b4f2b docstrings for checkpoint 2025-05-01 13:29:51 -04:00
Sydney Runkle 5916dc333e docstring fixes for cli 2025-05-01 13:23:54 -04:00
Sydney Runkle 7b749f05bd docstring fixes for libs/langgraph 2025-05-01 13:22:49 -04:00
Vadym BardaandGitHub cc4d2d26c7 docs: update add_messages API ref (#4495) 2025-05-01 12:33:27 -04:00
William FHandGitHub a2035eeb11 Update CLI (#4488) 
Remove maxbound for langgraph-api and runtime inmem.

Update min-bound.

Add server log level flag for dev
 2025-04-30 18:23:41 -07:00
William FHandGitHub 9a5b602287 Add checkpoint_during for the SDKs (#4487) 2025-04-30 15:56:29 -07:00
lc-arjunandGitHub 03c34bf2cf feat: assistants sorting sdk spec (#4484) 2025-04-30 17:05:37 -04:00
David DuongandGitHub 0269dd8818 release(langgraph): 0.4.1 (#4480) 2025-04-30 20:24:48 +02:00
Tat Dat Duong 91e9f12b54 release(sdk-js): 0.0.72 2025-04-30 20:19:30 +02:00
Tat Dat Duong 5779d9079f release(langgraph): 0.4.1 2025-04-30 20:18:25 +02:00
David DuongandGitHub e02c4b06db feat(ui): add merge option to UI messages (#4473) 
- Add docs about (partial) streaming UI components from LLMs
- Add missing support for "nostream" in LangGraph
 2025-04-30 20:15:57 +02:00
Tat Dat Duong be1af772f5 Fix lint 2025-04-30 20:04:48 +02:00
Tat Dat Duong b3371a1d63 Add missing "nostream" support 2025-04-30 19:50:14 +02:00
Tat Dat Duong 5804e788d8 Add docs 2025-04-30 19:46:14 +02:00
Sydney RunkleandGitHub 49c10a9188 packaging: removing pydantic v1 support (#4448) 
Also moving over any logic from `langchain-core` to here as we slowly
drop `langchain-core` dependency.

Pydantic v1 is no longer undergoing active maintenance and v2 has been
out for almost 2 years, so it seems like an appropriate time to drop v1
scar tissue.
 2025-04-30 09:10:47 -04:00
Tat Dat Duong 1b6e12ef14 feat(ui): add merge option to UI messages 2025-04-30 11:52:47 +02:00
Tat Dat Duong 20b9b8b1d7 release(sdk): Python 0.1.64 & JS (0.0.71) 2025-04-30 10:33:52 +02:00
William Fu-Hinthorn 91de85a8e6 fix test 2025-04-29 15:55:54 -07:00
Tat Dat Duong d0a70a15a8 fix(sdk-js): avoid sending run metadata in UI messages 2025-04-30 00:34:22 +02:00
Sydney Runkle ef1e1659c6 conditional for config 2025-04-29 18:22:18 -04:00
Sydney Runkle c712e09fbc conditional for config 2025-04-29 18:17:33 -04:00
Vadym BardaandGitHub e25dde1df0 docs(reference): filter class methods and add missing docstrings (#4463) 2025-04-29 21:31:57 +00:00
David DuongGitHubNuno Campos
 		d86d0a9311 fix(langgraph): missing conditional edge on get_graph() (#4458) 
Co-authored-by: Nuno Campos <nuno@langchain.dev>
 2025-04-29 19:49:48 +00:00
Sydney Runkle 80dca9b9a5 removing remaining v1 logic 2025-04-29 15:11:37 -04:00
Vadym BardaandGitHub ea55c2d468 docs: simplify docstring / API reference for create_react_agent (#4457) 2025-04-29 18:52:43 +00:00
Sydney Runkle da0994b741 removing langchain-core pydantic utilities 2025-04-29 14:31:28 -04:00
Sydney Runkle 64cfbb0d02 remove v1 test 2025-04-29 13:51:45 -04:00
David DuongandGitHub 9ffa6371e0 feat(sdk-js): add onLangChainEvent and onDebugEvent to useStream (#4453) 2025-04-29 15:40:09 +02:00
David DuongandGitHub 4d1e5ab71f feat(sdk-js): pass the client instead of apiKey/apiUrl (#4452) 2025-04-29 15:38:47 +02:00
Tat Dat Duong 8ad5331c89 feat(sdk-js): add onLangChainEvent and onDebugEvent to useStream 2025-04-29 15:34:03 +02:00
Tat Dat Duong cc62a9fa33 feat(sdk-js): pass the client instead of apiKey/apiUrl 2025-04-29 15:17:55 +02:00
Sydney Runkle e99028cfc6 what would it look like to remove pydantic v1 support? 2025-04-28 21:44:17 -04:00
Sydney Runkle 6250b364f7 version bumps and locks 2025-04-28 17:39:20 -04:00
Sydney Runkle 2bf8e690b0 test + bug fix 2025-04-28 15:41:37 -04:00
Nuno CamposandGitHub 78581b80c2 Support multiple resume values with Command.resume (#4406) 
* Adding support for mapping interrupt ids -> resume values with the
`Command.resume_map` argument, like:

```py
resume_map = {
    i.interrupt_id: f"human input for prompt {i.value}" 
    for i in parent_graph.get_state(thread_config).interrupts 
}

parent_graph.invoke(Command(resume=resume_map), config=thread_config)
```

* Adds an `interrupts` attribute on `StateSnapshot` so that we can
access that directly rather than having to do
`get_state(thread_config).tasks` and then iterate over tasks to find
interrupts

* Deprecates undocumented feature where (if interrupting a graph from
the level of an interrupt), you could pass a dict mapping task ids ->
resume values. Now we recommend and endorse the `interrupt_id` approach
above.

I'll note, from an internal perspective, I would love if we didn't have
to pass around this map, but it seems like the best way right now to
make the necessary resume information necessary at different levels in a
graph with subgraphs.

Fix https://github.com/langchain-ai/langgraph/issues/4028

Slotted to be included in our v0.4.0 release early next week!
 2025-04-28 09:06:58 -07:00