## langchain-ai/langchain — v0.1.14…v0.1.15

_146 commits._

### Features
- **community[minor]: add allow_dangerous_requests for OpenAPI toolkits (#19493)** (0394c6e)
- **community[minor]: add neptune analytics graph (#20047)** (cd7abc4)
- **community[minor]: add bedrock anthropic callback for token usage counting (#19864)** (ad97504)
- **community[minor]: Add support for MLX models (chat & llm) (#18152)** (1f9f4d8)
- **community: add request_timeout and max_retries to ChatAnthropic (#19402)** (a812839)
- **community: extend Predibase integration to support fine-tuned LLM adapters (#19979)** (5f563e0)
- **community: Add PHP language parser to document_loaders (#19850)** (e1a24d0)
- **community[minor]: Add support for Pebblo cloud_api_key in  PebbloSafeLoader (#19855)** (820b713)
- **postgres[minor]: Add pgvector community as is (#20096)** (34a24d4)
- **postgres[minor]: add postgres checkpoint implementation (#20025)** (ba9e0d7)
- **templates: migrate to langchain_anthropic package to support Claude 3 models (#19393)** (de49606)
- **templates: add rag azure search template (#18143)** (d64bd32)
- **core[minor]: Add async aformat_document method (#20037)** (7e5c190)
- **Add cookbook for Anthropic .with_structured_output() (#20017)** (a692677)
- **community[minor]: Add metadata filtering support for neo4j vector (#20001)** (df25829)
- **core[minor]: support pydantic V2 for JSONOutputParser, allow for other sources of JSON schemas (#19716)** (31e3ecc)
- **core[minor]: Add aformat to FewShotPromptTemplate (#19652)** (f97de4e)
- **groq: Add tool calling support (#19971)** (88cf8a2)
- **cli[minor]: Add disable sockets in unit tests (#19877)** (6f20f14)
- **core[minor]: Add aload to document loader (#19936)** (d293431)
- **community[minor]: add Layerup Security integration (#19787)** (4f70bc1)
- **core: Add docs for RunnableConfigurableFields (#19849)** (d5c412b)
- **text-splitters[minor]: Adding a new section aware splitter to langchain (#16526)** (c28efb8)
- **cli[minor]: Add __version__ to integration package template (#19876)** (356a139)
- **community[minor]: Add Dria retriever (#17098)** (4384fa8)
- **community[minor]: Add OpenVINO rerank model support (#19791)** (48f84e2)
- **community[minor]: Update ChatZhipuAI to support GLM-4 model (#16695)** (a1f3e9f)

### Fixes
- **anthropic[patch]: Release 0.1.7 (#20240)** (4b84c9b)
- **mistralai[patch]: Release 0.1.1 (#20239)** (74d04a4)
- **langchain[patch]: Release 0.1.15 (#20237)** (e5913c8)
- **community[patch]: Release 0.0.32 (#20236)** (e39fdfd)
- **core[patch]: Release 0.1.41 (#20233)** (a07238d)
- **community[patch]: fixed multithreading returning List[List[Documents]] instead of List[Documents] (#20230)** (806d4ae)
- **langchain[patch]: Update unit test (#20228)** (fe35e13)
- **langchain[patch]: make BooleanOutputParser check words not substrings (#20064)** (b972f39)
- **community[patch]: OpenLLM Async Client Fixes and Timeout Parameter (#20007)** (add31f4)
- **openai[patch]: wrap stream code in context manager blocks (#18013)** (a682f0d)
- **community[patch]: GCSDirectoryLoader bugfix (#20005)** (0c848a2)
- **community[patch], langchain[minor]: Enhance Tencent Cloud VectorDB, langchain: make Tencent Cloud VectorDB self query retrieve compatible (#19651)** (ac42e96)
- **community[patch]: pass through sql agent kwargs (#19962)** (1a34c65)
- **community[patch]: standardize baichuan init args (#20209)** (97d91ec)
- **community[patch]: `cross_encoders` flatten namespaces (#20183)** (2f8dd1a)
- **pinecone[patch]: fix core min version (#20177)** (391e8f2)
- **langchain: fix pinecone upsert when async_req is set to False (#19793)** (1ee2085)
- **anthropic[patch]: standardize init args (#20161)** (a27d88f)
- **mistralai[patch]: standardize model params (#20163)** (3490d70)
- **[core] fix: manually specifying run_id for chat models.invoke() and .ainvoke() (#20082)** (039b7a4)
- **pinecone[patch]: release 0.1.0 (#20109)** (96dc0ea)
- **partners[anthropic]: fix anthropic chat model message type lookup keys (#19034)** (5ac0d1f)
- **docs[patch]: Add missing redirects (#20076)** (58a2123)
- **community[patch]: Improve import callbacks to make it IDE friendly (#20050)** (520ff50)
- **core[patch]: Document BaseCache abstraction in code (#20046)** (e4fc0e7)
- **docs[patch]: Fix Model I/O quickstart (#20031)** (b69af26)
- **anthropic[patch]: Release 0.1.6 (#20026)** (1b7ed60)
- **anthropic[patch]: use anthropic 0.23 (#20022)** (6860450)
- **anthropic[patch]: fix experimental tests (#20021)** (be3dd62)
- **anthropic[patch]: bump core dep (#20019)** (86fdb79)
- **docs[patch]: Make Docusaurus and Vercel add trailing slashes when navigating by default (#20014)** (7f0cb3b)
- **[docs][minor]: Fix typo in Custom Document Loader doc (#20003)** (02152d3)
- **core[patch]: mypy ignore fixes #17048 (#19931)** (b27f81c)
- **cohere: Add citations to agent, flexibility to tool parsing, fix SDK issue (#19965)** (e103492)
- **groq: fix core version (#19976)** (5acb564)
- **core: fix try_load_from_hub for older langchain versions load_chain (#19964)** (83f62fd)
- **community[patch]: Implement delete method and all async methods in opensearch_vector_search (#17321)** (c6432ab)
- **core: fix return of draw_mermaid_png and change to not save image by default (#19950)** (31a641a)
- **core[patch]: Release 0.1.39 (#19940)** (4328c54)
- **core[patch]: remove requests (#19891)** (f0d5b59)
- **pinecone[patch]: source tag (#19739)** (d5a2ff5)
- **cohere: Improve integration test stability, fix documents bug (#19929)** (beab9ad)
- **cohere[patch]: release 0.1.0rc2 (#19924)** (146d1a6)
- **cohere[patch]: Add multihop tool agent (#19919)** (e2b83c8)
- **langchain[patch]: fix ElasticsearchStore reference for self query (#19907)** (22dbcc9)
- **core[patch]: Release 0.1.38 (#19895)** (3218463)
- **community[patch]: Add remove_comments option (default True): do not extract html comments (#13259)** (e830a4e)
- **docs[patch]: Hide google from function calling docs (#19887)** (22f78c3)
- **cohere[patch]: support request timeout in BaseCohere (#19641)** (06dac39)
- **docs[patch]: Revert quarto update (#19880)** (7538c4d)
- **robocorp[patch]: fix core min version (#19879)** (0b0a551)
- **robocorp[patch]: Fix nested arguments descriptors and tool names (#19707)** (3f06cef)
- **openai[patch]: remove openai chunk size validation (#19878)** (4fbdc2a)
- **community[patch]: Release 0.0.31 (#19873)** (d25b5b6)
- **ai21[patch]: fix core dep (#19874)** (e3ed6a7)
- **openai[patch]: Partially Revert Update openai chat model to new base class interface (#19871)** (aa5797d)
- **openai[patch]: fix azure embedding length check (#19870)** (be92cf5)
- **community[patch]: Revert " Fix the bug that Chroma does not specify `e… (#19866)** (d62e84c)
- **ai21[patch]: release 0.1.3 (#19867)** (7376e4d)
- **experimental[patch]: Release 0.0.56 (#19840)** (003c98e)

### Backend
- **community: switch to falkordb python client (#20229)** (37a9e23)
- **core[minor]: Implement aformat_messages for _StringImageMessagePromptTemplate (#20036)** (f43b48a)
- **core[minor]: Implement aformat for FewShotPromptWithTemplates (#20039)** (19001e6)
- **standard-tests: a standard unit and integration test set (#20182)** (855ba46)
- **together: release 0.1.0 (#20225)** (9b5cae0)
- **langchain-postgres: Remove remaining README.md file (#20221)** (7cfb643)
- **Remove postgres package (#20207)** (2fa7266)
- **cookbook: MongoDB Cookbook for Chat history and semantic cache (#19998)** (c769421)
- **Community: Updating Azure Retriever and Docs to be Azure AI Search instead of Azure Cognitive Search  (#19925)** (2f03bc3)
- **Documentation: Fixed the typo of Discord -> Telegram (#20008)** (ba602dc)
- **cohere: move package to external repo (#20081)** (28dfde2)
- **langchain-core[minor]: Allow passing local cache to language models (#19331)** (5a76087)
- **core[minor]: Implement aformat_prompt and ainvoke in BasePromptTemplate (#20035)** (4d8a6a2)
- **Merge pull request #20038** (927793d)
- **Update example cookbook for Anthropic tool use (#20029)** (e76b921)
- **docs `integrations/providers` update 10 (#19970)** (4c96928)
- **anthropic[minor]: tool use (#20016)** (209de0a)
- **community[minor]: added missed class to __all__ (#19888)** (3aacd11)
- **langchain[minor]: enhance `LocalFileStore` to allow directory/file permissions to be specified (#18857)** (a954ded)
- **community[minor]: Implement Async OpenSearch `afrom_texts` & `afrom_embeddings` (#20009)** (b52b784)
- **groq: handle streaming tool call case (#19978)** (51bdfe0)
- **groq: release 0.1.0 (#19975)** (9e60159)
- **langchain[minor]: Tests update metadata filtering examples of documents (#19963)** (09a0ecd)
- **community[minor]: weight only quantization with intel-extension-for-transformers. (#14504)** (cc407e8)
- **langchain-postgres: Initial package with postgres chat history implementation (#19884)** (d6d843e)
- **core: BaseChatModel modify chat message before passing to run_manager (#19939)** (f4568fe)
- **deprecating integrations moved to langchain_google_community (#19841)** (eb05210)
- **cohere: simplify integration test (#19928)** (37fc1c5)
- **cohere, docs: update imports and installs to langchain_cohere (#19918)** (de6c0cf)
- **Core[major]: Base Tracer to propagate raw output from tool for on_tool_end (#18932)** (9ae2df3)
- **core: Assign missing message ids in BaseChatModel (#19863)** (2ae6dcd)
- **👥 Update LangChain people data (#19858)** (f06229b)
- **core: generate mermaid syntax and render visual graph (#19599)** (c2ccf22)
- **Update cross_encoder_reranker.ipynb (#19846)** (8711a05)

### Docs
- **docs: Fix typo in citations example (#20218)** (230376f)
- **docs: Add documentation of `ElasticsearchStore.BM25RetrievalStrategy` (#20098)** (6c11c8d)
- **docs: Get rid of ZeroShotAgent and use create_react_agent instead (#20157)** (301dc3d)
- **docs: Fix the class links in openai_tools and openai_functions description in output parser documentations (#20197)** (1b48091)
- **docs: TogetherAI as a drop-in replacement for OpenAI (#19900)** (6baeaf4)
- **docs: add vertexai to structured output (#20171)** (1af7133)
- **docs: standardize fireworks params (#20162)** (1718240)
- **docs: use standard openai params (#20160)** (5ae0e68)
- **docs: anthropic tool docstring (#20091)** (46f580d)
- **docs: fix title cap (#20048)** (ebd24bb)
- **Docs: Update custom chat model (#19967)** (1ee8cf7)
- **docs: weaviate docs (#20042)** (5fc7bb0)
- **docs: fix together model tab (#20032)** (38fb142)
- **docs: fixing typo in argument name (#20028)** (94ac42c)
- **docs: hide experimental anthropic (#20030)** (07eeeb8)
- **docs: `integrations/providers` update 9 (#19941)** (3856ded)
- **docs: mark anthropic tools wrapper as deprecated (#20024)** (644ff46)
- **docs: `integrations/providers/unstructured` update (#19892)** (69bf626)
- **docs: `graphs` update (#19675)** (82f0198)
- **docs: reorg and visual refresh (#19765)** (605c3f2)
- **docs: Custom Document Loaders (#19935)** (ea276d6)
- **docs: update cohere documentation (#19700)** (73ebe78)
- **docs: mention caveats with CacheBackedEmbeddings.embed_query (#19926)** (8638029)
- **docs: Fix link in Unstructured notebook (#19851)** (dfbc10c)
- **docs: remove unnecessary args from the pip install (#19823)** (039f314)

_Recap by [Repo Wrapped](https://repowrapped.com/gh/langchain-ai/langchain?utm_source=github-action)._