langchain-ai/langchain - CodSpeed

langchain

Blog Docs Changelog

Performance History

Latest Results

refactor(typesafe): deprecate experimental middleware (#41768) Deprecate langchain-typesafe experimental middleware. | # | Change vs. langchain-typesafe 0.0.2 | Motivation | |---|---|---| | 1 | Router with no `HumanMessage` in state raises `ValueError` instead of leaking `StopIteration` | Fixing a bug | | 2 | `ModelRouterMiddleware.__init__` annotates `instructions` as `Content` instead of `_QuestionContent` | Uses type from langchain-core. The type is identical (`str \| dict[str, JsonValue] \| list[JsonValue]`) | | 3 | Update docstrings, readme | |

master

28 minutes ago

accommodate blockbuster

cc/decisions-middleware

38 minutes ago

update docstring in core

cc/decisions-core

38 minutes ago

update docstring in core

cc/decisions-core

17 hours ago

docs(langchain): explain Baseten minimum version (#41777) Explain that the Baseten extra requires the 0.2.0 `ChatBaseten` rewrite based on `BaseChatOpenAI`, rather than its legacy tool-calling and streaming implementation. ## References - [Slack thread](https://langchain.slack.com/archives/C0B4H54MB35/p1794283565322519) Made by [Robin](https://github.com/langchain-ai/open-swe) · [view thread](https://robin.langchain.dev/agents/d0003d7a-b3df-54aa-aa20-ec132fe765c4) · openai:gpt-6.1-sol (medium) Co-authored-by: Mason Daugherty <mdrxy@users.noreply.github.com> Co-authored-by: Robin <open-swe@users.noreply.github.com>

master

20 hours ago

release(langchain): 1.4.10 (#41778) Bump `langchain` to 1.4.10 and require `langchain-core>=1.7.7,<2.0.0` to include the latest deserialization fixes. Refresh lockfile metadata for local packages. Includes: - #41579: keep middleware-internal model calls off the `messages` stream. - #41689: omit `ModelRefusalMiddleware` trace inputs. - #41585: newer-Python dependency/security lockfile updates. ## References - [Slack thread](https://langchain.slack.com/archives/C0C5950ARJT/p1794282301267519) Made by [Robin](https://github.com/langchain-ai/open-swe) · [view thread](https://robin.langchain.dev/agents/a1aec474-a91d-5952-ab1e-9fca84ec15ff) · openai:gpt-6.1-sol (medium) Co-authored-by: Mason Daugherty <mdrxy@users.noreply.github.com> Co-authored-by: Robin <open-swe@users.noreply.github.com>

master

20 hours ago

perf(core): offload async stream chunk merging

adity982:perf/offload-async-stream-chunk-merge

22 hours ago

fix(core): only follow cancel __context__ on Python < 3.11 Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

kiewan/cancelled-run-status-tracer

22 hours ago

Latest Branches

0%

feat(langchain): add model router and auto mode middleware#41766

2 days ago

68956ee

cc/decisions-middleware

0%

feat(core): add decision models#41765

17 hours ago

9db3611

cc/decisions-core

+1%

perf(core): offload async stream chunk merging#41786

22 hours ago

835b093

adity982:perf/offload-async-stream-chunk-merge

© 2026 CodSpeed Technology

Home Terms Privacy Docs