Skip to content

Releases: langchain-ai/langchain

langchain-huggingface==0.1.2

31 Oct 18:57
06420de
Compare
Choose a tag to compare

Changes since langchain-huggingface==0.1.1

integrations[patch]: bump core to 0.3.15 (#27805)
partners/huggingface[patch]: fix HuggingFacePipeline model_id parameter (#27514)

langchain-groq==0.2.1

31 Oct 18:34
06420de
Compare
Choose a tag to compare

Changes since langchain-groq==0.2.0

integrations[patch]: bump core to 0.3.15 (#27805)
groq[patch]: update standard tests (#27744)
groq: fix unit test (#27660)
groq[patch]: support tool_choice=any/required (#27000)
core,openai,grow,fw[patch]: deprecate bind_functions, update chat mod… (#26584)

langchain-fireworks==0.2.5

31 Oct 18:33
06420de
Compare
Choose a tag to compare

Changes since langchain-fireworks==0.2.1

integrations[patch]: bump core to 0.3.15 (#27805)
multiple: rely on asyncio_mode auto in tests (#27200)

langchain-core==0.3.15

31 Oct 18:06
25a1031
Compare
Choose a tag to compare

Changes since langchain-core==0.2.42

core[patch]: update image util err msg (#27803)
core[patch]: Release 0.3.15 (#27802)
core[patch]: rm image loading (#27797)
Improvement [docs] Improve api docs (#27787)
core[patch]: Release 0.3.14 (#27764)
core[patch]: make get_all_basemodel_annotations public (#27761)
core[patch]: support bedrock converse -> openai tool (#27754)
Support message trimming on single messages (#27729)
core: JsonOutputParser UTF characters bug (#27306)
core[patch]: Improve type checking for the tool decorator (#27460)
core: test runner improvements (#27654)
core[patch]: improve performance of InMemoryVectorStore (#27538)
all: test 3.13 ci (#27197)
core[patch]: Release 0.3.13 (#27651)
core: remove mustache in extended deps (#27629)
core: add flake8-bandit (S) ruff rules to core (#27368)
core[patch]: fix repr and str for Serializable (#26786)
core[patch]: convert_to_openai_tool Anthropic support (#27591)
core: fix Image prompt template hardcoded template format (#27495)
core: release 0.3.12 (#27410)
docs, core: error messaging [wip] (#27397)
core[patch]: 0.3.11 release (#27403)
core[patch]: add convert_to_openai_messages util (#27263)
multiple: rely on asyncio_mode auto in tests (#27200)
core[patch]: Ignore ASYNC110 to upgrade to newest ruff version (#27229)
core[patch],langchain[patch],community[patch]: Bump version dependency of tenacity to >=8.1.0,!=8.4.0,<10 (#27201)
core: release 0.3.10 (#27209)
core[patch]: utils for adding/subtracting usage metadata (#27203)
core[patch]: support ValidationError from pydantic v1 in tools (#27194)
core: Add various ruff rules (#26836)
core: update make format to fix all autofixable things (#27174)
core: Add ruff rules for error messages (EM) (#26965)
core[minor]: Improve support for id in VectorStore (#26660)
core[patch]: Update AsyncCallbackManager to honor run_inline attribute and prevent context loss (#26885)
core[minor]: add **kwargs to index and aindex functions for custom vector_field support (#26998)
core[patch]: correct typo doc-string for astream_events method (#27108)
core[patch]: Release 0.3.9 (#27117)
core,integrations[minor]: Dont error on fields in model_kwargs (#27110)
core: fix batch race condition in FakeListChatModel (#26924)
core[patch]: exclude model cache from ser (#27086)
core[patch]: add UsageMetadata details (#27072)
core[patch]: Release 0.3.8 (#27046)
core[patch]: Support injected tool args that are arbitrary types (#27045)
core[patch]: fix init of RunnableAssign (#26903)
multiple: update docs urls to latest 2 (#26837)
core[patch]: improve index/aindex api when batch_size<n_docs (#25754)
core[patch]: Add kwargs to Runnable (#27008)
Merge pull request #26794
core[patch]: Release 0.3.7 (#27004)
core[patch]: fix structured prompt template format (#27003)
core: Add ruff rules for pycodestyle Warning (W) (#26964)
core[patch]: Add ruff rules for flake8-simplify (SIM) (#26848)
docs: Upgrade examples with RunnableWithMessageHistory to langgraph memory (#26855)
core: Add ruff rules for PIE (#26939)
core[fix]: using async rate limiter methods in async code (#26914)
core: Fix IndexError when trim_messages invoked with empty list (#26896)
core: release 0.3.6 (#26863)
core: Add ruff rules for comprehensions (C4) (#26829)
[Core] Add more interops tests (#26841)
[Core] Inherit tracing metadata & tags (#26838)
core: Run LangChainTracer inline (#26797)
core[patch]: On Chain Start Fix for Chain Class (#26593)
core[patch]: set default on Blob (#26787)
core: release 0.3.5 (#26737)
[LangChainTracer] Omit Chunk (#26602)
[Tracer] add project name to run from tracer (#26736)
core: fix "template" not allowed as prompt param (#26060)
core: release 0.3.4 (#26729)
core: In astream_events propagate cancellation reason to inner task (#26727)
core: release 0.3.3 (#26713)
core[patch]: Fix defusedxml import (#26718)
core/langchain: fix version used in deprecation (#26724)
core[patch],langchain[patch]: deprecate memory and entity abstractions and implementations (#26717)
Avoid copying runs (#26689)
core[patch]: support js chat model namespaces (#26688)
core[patch]: Release 0.3.2 (#26686)
core[patch]: Fix load of ChatBedrock (#26679)
core[patch]: support load from path for default namespaces (#26675)
core, community: move graph vectorstores to community (#26678)
core: Add N(naming) ruff rules (#25362)
core: use ruff.target-version instead (#26634)
core: remove [project] tag from pyproject (#26633)
core: Put Python version as a project requirement so it is considered by ruff (#26608)
core[patch]: Add ruff rule UP006(use PEP585 annotations) (#26574)
docs: pypi readme image links (#26590)
core,openai,grow,fw[patch]: deprecate bind_functions, update chat mod… (#26584)
core[patch]: Release 0.3.1 (#26581)
core[patch]: Fix "argument of type 'NoneType' is not iterable" error in LangChainTracer (#26576)
core[minor]: remove beta from RemoveMessage (#26579)
core: release 0.3.0 (#26453)
multiple: pydantic 2 compatibility, v0.3 (#26443)

langchain-core==0.2.43

31 Oct 20:30
9fdeb74
Compare
Choose a tag to compare

Changes since langchain-core==0.3.15

core[patch]: Release 0.2.43 (#27808)
core[patch]: remove prompt img loading (#27807)
core[patch]: Release 0.2.42 (#27763)
core[patch]: make get_all_basemodel_annotations public (#27762)
Added mapping to fix CI for #langchain-aws:227. (#27114)
core[patch]: Release 0.2.41 (#26687)
core[patch]: Fixed bedrock chat model load. (#26643)

langchain-community==0.3.4

31 Oct 00:53
128b072
Compare
Choose a tag to compare

Changes since langchain-community==0.3.3

community: release 0.3.4 (#27769)
community[patch]: Fix Playwright Tools bug with Pydantic schemas (#27050)
community: prevent gitlab commit on main branch for Gitlab tool (#27750)
community: Add Writer integration (#27646)
community: Update Replicate LLM and fix tests (#27655)
docs: Add OpenAIAssistantV2Runnable docstrings (#27402)
community: add new parameters to pass to OpenAIAssistantV2Runnable (#27372)
community: Add token_usage and model_name metadata to ChatZhipuAI stream() and astream() response (#27677)
community: Added anthropic.claude-3-5-sonnet-20241022-v2:0 cost detials (#27728)
community[patch]: Fix vLLM integration to apply lora_request (#27731)
community[pathch]: Add missing custom content_key handling in Redis vector store (#27736)
community: Fix closed session in Infinity (#26933)
community: sambastudio llm refactor (#27215)
all: test 3.13 ci (#27197)
community[patch]: Fix missing protected_namespaces(). (#27610)
Improvement[Community] Improve Document Loaders and Splitters (#27568)
community: Add Naver chat model & embeddings (#25162)
community: Fix the failure of ChatSparkLLM after upgrading to Pydantic V2 (#27418)
Community: Fix DuckDuckGo search tool Output Format (#27479)
partners/openai + community: Async Azure AD token provider support for Azure OpenAI (#27488)
community: Update file_path type in JSONLoader.__init__() signature (#27535)
community: Cassandra Vector Store: modernize implementation (#27253)
community: fix DallE hidden open_api_key (#26996)
community: fix a typo in planner_prompt.py (#27489)
community: fix lint from new mypy (#27474)

langchain-anthropic==0.2.4

31 Oct 18:35
06420de
Compare
Choose a tag to compare

Changes since langchain-anthropic==0.2.3

integrations[patch]: bump core to 0.3.15 (#27805)
anthropic[patch]: allow multiple sys not at start (#27725)
core[patch]: add convert_to_openai_messages util (#27263)
multiple: rely on asyncio_mode auto in tests (#27200)
[Anthropic] Shallow Copy (#27105)
docs: fix anthropic max_tokens docstring (#27166)

langchain==0.3.6

31 Oct 00:01
1ed3cd2
Compare
Choose a tag to compare

Changes since langchain==0.3.5

langchain: release 0.3.6 (#27768)

langchain==0.3.5

30 Oct 00:11
5111063
Compare
Choose a tag to compare

Changes since langchain==0.3.4

langchain[patch]: Release 0.3.5 (#27727)
docs: Fix typo in _action_agent docs section (#27698)
all: test 3.13 ci (#27197)
langchain[patch]: de-beta init_chat_model (#27558)

langchain-text-splitters==0.3.1

30 Oct 00:10
5111063
Compare
Choose a tag to compare

Changes since langchain-text-splitters==0.3.0

text-splitters[patch]: Release 0.3.1 (#27726)
all: test 3.13 ci (#27197)
Improvement[Community] Improve Document Loaders and Splitters (#27568)
multiple: rely on asyncio_mode auto in tests (#27200)
Improvement[text-splitter] Fix import of ExperimentalMarkdownSyntaxTextSplitter (#26703)