langchain-ai/langchain v0.1.17rc1

on GitHub

What's Changed

• docs: tool agent nit by @baskaryan in #20337
• docs: update tool-calling table by @baskaryan in #20338
• docs: Update chat model providers include package information by @eyurtsev in #20336
• chroma: remove relevance score int test by @efriis in #20346
• docs: update anthropic tool call by @baskaryan in #20344
• docs: use vertexai in chat model tabs by @baskaryan in #20352
• chroma: release 0.1.0 by @efriis in #20355
• docs: tool agent nit by @baskaryan in #20353
• docs: tool call nits by @baskaryan in #20356
• core[patch]: Update documentation for base retriever by @eyurtsev in #20345
• community: docstrings update by @leo-gan in #20301
• docs: add Cloud SQL for MySQL vector store integration docs by @jackwotherspoon in #20278
• docs: Update ibm_watsonx.ipynb by @eltociear in #20329
• community: add gpt-4 pricing in callback by @IsakNyberg in #20292
• docs: show tool msg in tool call docs by @baskaryan in #20358
• multiple: standard chat model tests by @efriis in #20359
• core: bind_tools interface on basechatmodel by @efriis in #20360
• Add another unit test for indexing code by @eyurtsev in #20387
• langchain[patch]: Add deprecation warning to extraction chains by @eyurtsev in #20224
• docs: add Cohere to ChatModelTabs by @ccurme in #20386
• community: for Predibase -- enable both Predibase-hosted and HuggingFace-hosted fine-tuned adapter repositories by @alexsherstinsky in #20370
• docs: vertexai tool call update by @baskaryan in #20362
• community[patch]: Add "model" attribute to the payload sent to Ollama in ChatOllama by @ptgoetz in #20354
• docs: Adding api documentation link at the end of each output parser class description page. by @Haris-Ali007 in #20391
• docs: langchain-chroma package by @efriis in #20394
• Skip pydantic validation for llm graph transformer and fix JSON response where possible by @tomasonjo in #19915
• community: Firecrawl.dev integration by @nickscamara in #20364
• community[patch]: Invoke callback prior to yielding token fix for HuggingFaceEndpoint by @balloonio in #20366
• Update index.mdx by @milovate in #20395
• community[patch]: Invoke callback prior to yielding token fix for Llamafile by @balloonio in #20365
• docs: fix extraction/quickstart.ipynb example code by @m-atoms in #20397
• templates: readme langsmith not private beta by @efriis in #20173
• langchain: callbacks imports fix by @leo-gan in #20348
• partner[ai21]: masking of the api key for ai21 models by @sepiatone in #20257
• Docs: Add guide for implementing custom retriever by @eyurtsev in #20350
• core[patch]: configurable init params by @baskaryan in #20070
• community[patch]: Databricks - fix scope of dangerous deserialization error in Databricks LLM connector by @dbczumar in #20368
• community[patch]: update use of deprecated llm methods by @ccurme in #20393
• docs: tutorials update by @leo-gan in #20401
• Make Groq .stream robust to ToolMessage by @ccurme in #20417
• text-splitters:Complete lua support in langchain.text_splitter module by @saberuster in #20421
• changed the link to more helpful source by @kennethchoe in #20411
• community: Set default value for _ListSQLDatabaseToolInput tool_input by @os1ma in #20409
• docs: Update Portkey provider integration by @roh26it in #20412
• community:sparkllm[patch]: standardize init args by @liugddx in #20194
• community: Fix YandexGPT embeddings by @egorkrash in #19720
• core: Refine RunnablePassthrough docstring by @smalyu in #19812
• community: Add Datahareld tool by @jcjc712 in #19680
• community[patch]: Invoke callback prior to yielding token fix [DeepInfra] by @balloonio in #20427
• community[patch]: deprecating remaining google_community integrations by @lkuligin in #20471
• docs: Add integration doc for postgres vectorstore by @eyurtsev in #20473
• Fix CDN URL in mermaid graph renderer by @angeligareta in #20381
• docs: quickstart retrieval chain for Cohere(API) by @NoumaanAhamed in #20475
• Remove logic to exclude intermediate nodes from rendering time by @angeligareta in #20459
• docs: tutorials update by @leo-gan in #20483
• docs: changed model's name in google-vertex-ai integration to a publicly available model by @lkuligin in #20482
• docs: Add Google Firestore Vectorstore doc by @averikitsch in #20078
• core: Add concept of conditional edge to graph rendering by @nfcampos in #20480
• core: forward config params to default by @efriis in #20402
• core: release 0.1.43 by @efriis in #20489
• community: release 0.0.33 by @efriis in #20490
• docs: update tool use case by @baskaryan in #20404
• community[patch]: Use uuid4 not uuid1 by @eyurtsev in #20487
• community: Fix pgvector deprecated filter clause usage with OR and AND conditions by @MartinGotelli in #20446
• anthropic: env ANTHROPIC_API_URL not work by @fayfox in #20507
• core[patch]: Fix imports defined in messages sub-package by @djbauer2 in #20500
• Add glue catalog loader by @ravediamond in #20220
• standard-tests: fix default fixtures by @efriis in #20520
• core[minor], langchain[patch]: tools dependencies refactoring by @leo-gan in #18759
• docs: response metadata by @baskaryan in #20527
• community[minor]: add UpTrainCallbackHandler by @Dominastorm in #19956
• Fix getattr in runnable binding for cases where config is passed in as arg too by @nfcampos in #20528
• Runnable graph viz improvements by @nfcampos in #20529
• update scheduled tests by @ccurme in #20526
• core: default chat/llm _identifying_params to lc_attributes by @efriis in #20232
• anthropic[patch]: use tool calls if content is empty by @baskaryan in #20535
• anthropic[patch]: bump to core 0.1.43 by @baskaryan in #20537
• anthropic[patch]: serialization in partner package by @efriis in #18828
• anthropic: support tool messages by @ccurme in #20519
• langchain-community: Make ChatDatabricks model supports streaming response by @WeichenXu123 in #19912
• community: Added integrations for ThirdAI's NeuralDB as a Retriever by @benitoThree in #17334
• docs: add snowflake provider page by @efriis in #20538
• community: update OctoAI endpoint to subclass BaseOpenAI by @sfvaroglu in #19757
• community: Changes to base_o365 and sharepoint document loaders by @MacanPN in #20373
• community: Improve Kuzu Cypher generation prompt by @prrao87 in #20481
• community: Fix not to cast to str type when file_path is None by @kozistr in #20057
• community: Added VLite as VectorStore by @sdan in #20245
• tongyi[patch]: standardize init args by @liugddx in #20210
• community:(baidu_qianfan)[patch]: standardize init args by @liugddx in #20166
• community: Add Documentation to enable authorized access identities in GoogleDriveLoader. by @rahul-trip in #20065
• community: Add support for authorized identities in PebbloSafeLoader. by @rahul-trip in #20055
• Community: Unify Titan Takeoff Integrations and Adding Embedding Support by @pjb157 in #18775
• docs: add structred output to feat table by @baskaryan in #20539
• Fix #18782 (Add streaming logic in ChatHuggingFace) by @eunhye1kim in #18784
• community: add ChatOctoAI by @sfvaroglu in #20059
• mistralai[patch]: Surface http errors by @eyurtsev in #20555
• mistral: read tool calls from AIMessage by @ccurme in #20554
• core[patch]: Release 0.1.44 by @baskaryan in #20564
• anthropic[patch]: Handle empty text block by @eyurtsev in #20566
• mistral, openai: allow anthropic-style messages in message histories by @ccurme in #20565
• doc(agent_with_memory_in_db):Get rid of ZeroShotAgent and use create_react_agent instead by @liugddx in #20155
• doc(agent_with_memory):Get rid of ZeroShotAgent and use create_react_agent instead by @liugddx in #20154
• partner(zep): implement add_messages and aadd_messages by @kaijietti in #20099
• community: Add function call support in Tongyi chat model. by @pcliupc in #20119
• anthropic[patch]: Release 0.1.10 by @baskaryan in #20568
• community: fix DirectoryLoader progress bar by @tomercagan in #19821
• community: Improve CassandraVectorStore from_texts by @cbornet in #20284
• community: Replace function in CassandraVectorStore with simpler lambda by @cbornet in #20323
• Lock low level dep to fix Vercel docs build by @jacoblee93 in #20573
• core: fix attributeerror in runnablelambda.deps by @nfcampos in #20569
• anthropic[patch]: fix msg mutation by @baskaryan in #20572
• exposing headless flag parameter to AsyncChromiumLoader class by @PeriniM in #20424
• upstage: init package by @efriis in #20574
• docs: aws docs updates by @efriis in #20571
• upstage: fix core dep by @efriis in #20576
• docs: nits by @efriis in #20577
• community, core[callbacks]: move FileCallbackHandler from community to core by @sepiatone in #20495
• docs[callbacks]: update to the FileCallbackHandler documentation by @sepiatone in #20496
• Support to create shards_num in milvus vectorstores by @rgupta2508 in #20318
• community: (milvus) check for num_shards by @ccurme in #20603
• fix typo in langchain/docs/docs/intergrations/tools/nuclia.ipynb by @Justsosostar in #20591
• Update RAPTOR.ipynb by @eltociear in #20586
• Update README.md -- broken hyperlink by @hodgesd in #20422
• update zhipuai notebook by @zRzRzRzRzRzRzR in #20595
• community: Add save_model function for openvino reranker and embedding by @OpenVINO-dev-contest in #19896
• community[patch]: Invoke callback prior to yielding token fix [HuggingFaceTextGenInference] by @balloonio in #20426
• core: fix fireworks mapping by @efriis in #20613
• mistralai[patch]: Use public attribute for eventsource.response by @eyurtsev in #20580
• community[minor]: Add async methods to CassandraLoader by @cbornet in #20609
• community: Add hybrid search to Cassandra VectorStore by @cbornet in #20286
• docs: import update by @leo-gan in #20610
• fix(community): AzureSearch incorrectly converted to retriever by @mspronesti in #20601
• experimental[patch]: prompts import fix by @leo-gan in #20534
• community[minor]: Update uptrain.ipynb to show outputs by @Dominastorm in #20551
• docs: Remove example vsdx data by @eyurtsev in #20620
• mistral[patch]: Support both model and model_name by @eyurtsev in #20557
• Ignore null and invalid embedding values for neo4j metadata filtering by @tomasonjo in #20558
• docs: Update body_func to hybrid_query in ElasticsearchRetriever by @naaive in #20498
• comminuty[patch]:Fix Error in apache doris insert by @liugddx in #19989
• Community : Add audio-parser "faster-whisper" in audio.py by @hulitaitai in #20012
• core: deprecation default to qualname by @efriis in #20578
• infra: fix doc imports ci by @efriis in #20629
• community[patch]: Fix Ollama for LLaMA3 by @rlancemartin in #20624
• community[patch]: truncate zhipuai temperature and top_p parameters to [0.01, 0.99] by @Congyuwang in #20261
• community: update Replicate to work with official models by @cbh123 in #20633
• Fix runnable map ser/de by @nfcampos in #20631
• upstage[patch]: un-xfail tool calling test, release 0.1.0 by @efriis in #20635
• infra: add test-doc-imports to ci failure by @efriis in #20637
• openai: fix allowed block types by @ccurme in #20636
• langchain[minor]: Databricks vector search self query integration by @s-udhaya in #20627
• docs: Update openvino example documents links by @OpenVINO-dev-contest in #20638
• docs: imports update by @leo-gan in #20625
• docs: providers alibaba update by @leo-gan in #20560
• docs: integrations/retrievers cleanup by @leo-gan in #20357
• docs: fix variable name typo in example code by @Souls-R in #20658
• docs: Fix minor typo in data_connection/document_loaders/custom by @djboris9 in #20648
• Updated Tutorials for Vertex Vector Search by @Adi8885 in #20376
• docs: Remove Duplicate --quiet Flag in Installation Command in LangSmith Docs by @saurabhchalke in #20121
• core: release 0.1.45 by @efriis in #20664
• Add support for relationship indexes in neo4j vector by @tomasonjo in #20657
• Update neo4j vector documentation by @tomasonjo in #20455
• Update neo4j cypher templates to the function callback by @tomasonjo in #20515
• community: release 0.0.34 by @efriis in #20672
• ibm: Add support for Embedding Models by @MateuszOssGit in #20647
• docs: aws listing by @efriis in #20674
• community[patch]: YandexGPT API add ability to disable request logging by @tyumentsev4 in #20670
• templates: utilities import fix by @leo-gan in #20679
• langchain: dosctore import fix by @leo-gan in #20678
• langchain: example_selector import fix by @leo-gan in #20676
• community: Add async methods to CassandraVectorStore by @cbornet in #20602
• community: add graph store implementation for apache age by @shumway743 in #20582
• add docs on google-genai vs google-vertexai by @lkuligin in #20713
• langchain[patch]: langchain-pinecone self query support by @baskaryan in #20702
• community[patch], experimental[patch]: support tool-calling sql and p… by @baskaryan in #20639
• community: Fix the stop sequence key name for Mistral in Bedrock by @matheus-rdo in #20709
• docs; Fixed grammar mistake by @leftmove in #20697
• patch: deprecate (a)get_relevant_documents by @ccurme in #20477
• Community patch clickhouse make it possible to not specify index by @mneedham in #20460
• community[patch],core[minor]: Move BaseToolKit to core.tools by @eyurtsev in #20669
• langchain[patch]: Add missing deprecation for openai adapters by @eyurtsev in #20668
• docs: automatically add api ref links by @baskaryan in #20755
• infra: rm sep repo partner dirs by @baskaryan in #20756
• community: Remove usage of deprecated StoredBlobHistory in CassandraChatMessageHistory by @cbornet in #20666
• docs: install partner pkgs vercel by @baskaryan in #20761
• docs: Fix link for partition_pdf in Semi_Structured_RAG.ipynb cookbook by @taabishm2 in #20763
• docs: update Memgraph docs by @katarinasupe in #20736
• Update google_drive.ipynb by @monke111 in #20731
• Bagatur/partner pkgs vercel by @baskaryan in #20766
• docs: update examples in api ref by @baskaryan in #20768
• Update custom_retriever.ipynb by @Chen94yue in #20776
• core, openai: support custom token encoders by @ccurme in #20762
• core[minor],langchain[patch],community[patch]: Move InMemory and File implementations of Chat History to core by @eyurtsev in #20752
• experimental: upgrade the prompt injection model by @asofter in #20783
• core[patch]: Remove autoupgrade to addable dict in Runnable/RunnableLambda/RunnablePassthrough transform by @eyurtsev in #20677
• Fix list rendering by @stefano-ottolenghi in #20774
• community[patch],core[minor]: Move in memory cache implementation to core by @eyurtsev in #20753
• langchain[major]: breaks some chains to remove hidden defaults by @eyurtsev in #20759
• community[patch]: fix CometTracer bug by @alexkuzmik in #20796
• core: use qualname in beta message by @efriis in #20361
• langchain: deprecate LLMChain, RetrievalQA, and ConversationalRetrievalChain by @ccurme in #20751
• langchain[major]: Remove default instantations of LLMs from VectorstoreToolkit by @eyurtsev in #20794
• langchain[patch]: Update evaluation logic that instantiates a default LLM by @eyurtsev in #20760
• mistral: catch GatedRepoError, release 0.1.3 by @ccurme in #20802
• standard-tests: split tool calling test by @efriis in #20803
• standard tests: add xfails by @ccurme in #20659
• groq: fix warning filter for integration test by @ccurme in #20806
• community[patch]: Fix message formatting for Anthropic models on Amazon Bedrock by @Neet-Nestor in #20801
• groq: release 0.1.3 by @efriis in #20836
• groq: bump core by @ccurme in #20838
• core,community: Move file chat history back to community by @eyurtsev in #20834
• groq: handle null queue_time by @ccurme in #20839
• core[minor],langchain[patch]: Move base indexing interface and logic to core by @eyurtsev in #20667
• docs: contributing note by @baskaryan in #20843
• feat(community): add support for similarity_score_threshold search in… by @mspronesti in #20852
• docs: Update Zep Messaging, add links to Zep Cloud Docs by @paul-paliychuk in #20848
• perplexity[patch]: standardize chat init args by @JeffKatzy in #20844
• docs: update openai llm documentation by @sepiatone in #20827
• Better support for subgraphs in graph viz by @nfcampos in #20840
• docs, multiple: de-beta with_structured_output by @efriis in #20850
• community: Support both Predibase SDK-v1 and SDK-v2 in Predibase-LangChain integration by @alexsherstinsky in #20859
• community: Advanced filtering for HANA Cloud Vector Engine by @MartinKolbAtWork in #20821
• community: tigergraph fixes by @leo-gan in #20034
• Add driver config param for neo4j graph by @tomasonjo in #20772
• community: add hybrid search to singlestoredb vectorstore by @volodymyr-memsql in #20793
• FIX: groq proxy support by @back2nix in #20758
• support messages in messages out by @hwchase17 in #20862
• langchain: add aprep_output method to langchain/chains/base.py by @hslee16 in #20748
• docs: imported List is not used by @hsmtkk in #20720
• partner: Add Upstage partner package LA and GC by @junkeon in #20651
• upstage: release 0.1.1 by @efriis in #20864
• Update apache_age.ipynb by @dpdjvhxm in #20722
• docs: [Retrieval > .. > PDF] update package installation instructions for Unstructured and PDFMiner by @salikadave in #20723
• langchain: fix agent_token_buffer_memory not working with openai tools by @pokidyshev in #20708
• community: LanceDB integration patch update by @raghavdixit99 in #20686
• patch: remove usage of llm, chat model call by @ccurme in #20788
• fix: git dependency name correction by @jtanios in #20662
• community: fix MarkdownHeaderTextSplitter fails to parse headers with non-printable characters by @coolbeevip in #20645
• [Core] Feat: update config CVar in tool.invoke by @hinthornw in #20808
• experimental: Makes regex customizable in text_splitter.py (SemanticChunker class) by @GustavoSept in #20485
• infra: fix how Poetry is installed in the dev container by @ibratoev in #20521
• mistral: comment batching param by @efriis in #20868
• docs[callbacks]: custom callback handlers page by @sepiatone in #20494
• community: added Browserbase loader by @mishushakov in #20478
• core[patch]: Release 0.1.46 by @baskaryan in #20891
• partner: Upstage quick documentation update by @chosh0615 in #20869
• core[minor], langchain[patch], community[patch]: mv StructuredQuery by @baskaryan in #20849
• langchain_openai: Allow disablling safe_len_embeddings(OpenAIEmbeddings) by @mokeyish in #19743
• community: add BeautifulSoupTransformer remove_unwanted_classnames method by @JasonSTong in #20467
• multiple: remove external repo mds by @efriis in #20896
• langchain: add Jina Reranker in retrievers module by @JoanFM in #19406
• docs: Fix misplaced zep cloud example links by @paul-paliychuk in #20867
• upstage: release 0.1.2 by @efriis in #20898
• docs: Use voyage-law-2 in the examples by @fzowl in #20784
• docs: Fix broken link in agents.ipynb by @samanhappy in #20872
• community: add HTTP response headers Content-Type to metadata of RecursiveUrlLoader document by @coolbeevip in #20875
• Support passing graph object to Neo4j integrations by @tomasonjo in #20876
• docs: remove unnecessary code snippet in multiple_tools.ipynb by @merdan-9 in #20883
• core, community: deprecate tool.call by @ccurme in #20900
• docs: update chat model feature table by @ccurme in #20899
• mistral, openai: support custom tokenizers in chat models by @ccurme in #20901
• langchain_community: deprecate persist method in Chroma by @AndresAlgaba in #20855
• community: fix tqdm import by @davidefantiniIntel in #20263
• community: Add relyt vector database by @klaus-xiong in #20316
• cli[minor]: Add version by @eyurtsev in #20903
• Add semantic info to metadata, classified by pebblo-server. by @rahul-trip in #20468
• [community] add more data types support to ipex-llm llm integration by @shane-huang in #20833
• experimental: remove \n from AutoGPT feedback_tool exit check by @mjschock in #20132
• core: improve comma separated list output parser to handle non-space separated list by @anish749 in #20434
• experimental: return from HuggingGPT task executor task.run() exception by @mjschock in #20219
• Implemented Kinetica Document Loader and added notebooks by @am-kinetica in #20002
• community[minor]: Add support for pebblo server and client version by @dristysrivastava in #20269
• community: Add initial tests for AzureSearch vector store by @mattgotteiner in #17663
• langchain: support PineconeVectorStore in self query retriever by @ccurme in #20905
• Use lstv2 by @hinthornw in #20747
• core: remove placeholder error message by @ccurme in #20340
• community[patch]: fix integrated test case test_recursive_url_loader.py assertions (issue-20919) by @coolbeevip in #20920
• docs: add tool calling example in Tongyi chat model integration. by @pcliupc in #20925
• cli[minor]: Add first version of migrate by @eyurtsev in #20902
• openai: add unit test by @ccurme in #20931
• cli[minor]: Add script to generate migrations for partner packages by @eyurtsev in #20932
• cli[minor]: Improve partner migrations by @eyurtsev in #20938
• openai: release 0.1.4 by @ccurme in #20939
• mistral: release 0.1.4 by @ccurme in #20940
• upstage: Upstage Groundedness Check parameter update by @chosh0615 in #20914
• upstage: release 0.1.3 by @efriis in #20941
• langchain: deprecate create_structured_output_runnable by @ccurme in #20933
• step1. adding langchain_google_community to the docs by @lkuligin in #20665
• infra: nonmaster release checkbox by @efriis in #20945
• cli[minor]: update code to generate migrations from langchain to community by @eyurtsev in #20946
• (infra) pass input to test-release by @ccurme in #20947

New Contributors

• @ptgoetz made their first contribution in #20354
• @balloonio made their first contribution in #20366
• @milovate made their first contribution in #20395
• @m-atoms made their first contribution in #20397
• @dbczumar made their first contribution in #20368
• @saberuster made their first contribution in #20421
• @roh26it made their first contribution in #20412
• @egorkrash made their first contribution in #19720
• @smalyu made their first contribution in #19812
• @jcjc712 made their first contribution in #19680
• @NoumaanAhamed made their first contribution in #20475
• @MartinGotelli made their first contribution in #20446
• @fayfox made their first contribution in #20507
• @djbauer2 made their first contribution in #20500
• @ravediamond made their first contribution in #20220
• @Dominastorm made their first contribution in #19956
• @WeichenXu123 made their first contribution in #19912
• @MacanPN made their first contribution in #20373
• @pjb157 made their first contribution in #18775
• @pcliupc made their first contribution in #20119
• @tomercagan made their first contribution in #19821
• @PeriniM made their first contribution in #20424
• @Justsosostar made their first contribution in #20591
• @hodgesd made their first contribution in #20422
• @zRzRzRzRzRzRzR made their first contribution in #20595
• @naaive made their first contribution in #20498
• @Congyuwang made their first contribution in #20261
• @s-udhaya made their first contribution in #20627
• @Souls-R made their first contribution in #20658
• @djboris9 made their first contribution in #20648
• @saurabhchalke made their first contribution in #20121
• @shumway743 made their first contribution in #20582
• @matheus-rdo made their first contribution in #20709
• @leftmove made their first contribution in #20697
• @mneedham made their first contribution in #20460
• @taabishm2 made their first contribution in #20763
• @monke111 made their first contribution in #20731
• @Chen94yue made their first contribution in #20776
• @stefano-ottolenghi made their first contribution in #20774
• @Neet-Nestor made their first contribution in #20801
• @paul-paliychuk made their first contribution in #20848
• @JeffKatzy made their first contribution in #20844
• @back2nix made their first contribution in #20758
• @hslee16 made their first contribution in #20748
• @hsmtkk made their first contribution in #20720
• @junkeon made their first contribution in #20651
• @dpdjvhxm made their first contribution in #20722
• @salikadave made their first contribution in #20723
• @jtanios made their first contribution in #20662
• @coolbeevip made their first contribution in #20645
• @GustavoSept made their first contribution in #20485
• @ibratoev made their first contribution in #20521
• @mishushakov made their first contribution in #20478
• @chosh0615 made their first contribution in #20869
• @JasonSTong made their first contribution in #20467
• @merdan-9 made their first contribution in #20883
• @AndresAlgaba made their first contribution in #20855
• @davidefantiniIntel made their first contribution in #20263
• @klaus-xiong made their first contribution in #20316
• @mjschock made their first contribution in #20132
• @anish749 made their first contribution in #20434
• @dristysrivastava made their first contribution in #20269

Full Changelog: v0.1.16...v0.1.17rc1