Releases: langchain-ai/langchain
langchain-ollama==0.1.3
Changes since langchain-ollama==0.1.2
ollama[patch]: Release 0.1.3 (#25902)
langchain-core==0.2.37
langchain-community==0.2.15
Changes since langchain-community==0.2.14
community[patch]: bump + fix core dep (#25901)
community: Fix branch not being considered when using GithubFileLoader (#20075)
community: Fix a bug in handling kwargs overwrites in Predibase integration, and update the documentation. (#25893)
qdrant: Updated class check in Self-Query Retriever factory (#25877)
community: Fixed infinity embeddings async request (#25882)
community:Fix a bug of LLM in moonshot (#25878)
Improvement[Community] Add tool-calling test case for ChatZhipuAI (#25884)
add neo4j query constructor for self query (#25288)
"community: release 0.2.15" (#25867)
community: undo azure_ad_access_token breaking change (#25818)
community: Cap AzureOpenAIEmbeddings chunk_size at 2048 instead of 16. (#25852)
added octoai test (#21793)
premai[patch]: Standardize premai params (#21513)
community(sparkllm): Add function call support in Sparkllm chat model. (#20607)
Community: sambastudio embeddings GenericV2 API support (#25064)
community[patch]: callback before yield for _astream (gigachat) (#25834)
Patch enable to use Amazon OpenSearch Serverless(aoss) for Semantic Cache store (#25833)
comminity[patch]: fix #25575 YandexGPTs for _grpc_metadata (#25617)
community[patch]: Support passing extra params for executing functions in UCFunctionToolkit (#25652)
community[patch]: optimize xinference llm import (#25809)
community[patch]: Add ID field back to Azure AI Search results (#25828)
langchain-prompty==0.0.3
Changes since langchain-prompty==0.0.2
prompty: bump core version (#25831)
prompty: release 0.0.3 (#25830)
community: avoid double templating in langchain_prompty (#25777)
[Community]: Added Template Format Parameter in create_chat_prompt for Langchain Prompty (#25739)
all: Improve make lint command (#25344)
all: add release notes to pypi (#24519)
prompty: docstring (#23152)
docs: fix some spelling mistakes caught by newest version of code spell (#22090)
langchain==0.2.15
Changes since langchain==0.2.14
langchain[patch]: Release 0.2.15 (#25802)
date_time_parser (#25763)
all: Improve make lint command (#25344)
docs: api ref to new site somewheres (#25679)
core: pydantic output parser streaming fix (#24415)
langchain: Fix type warnings when passing Runnable as agent to AgentExecutor (#24750)
fix issue: cannot use document_variable_name to override context in create_stuff_documents_chain (#25531)
docs: fix Agent deprecation msg (#25464)
[experimental]: minor fix to open assistants code (#24682)
langchain[patch]: deprecate various chains (#25310)
langchain-openai==0.1.23
Changes since langchain-openai==0.1.22
openai[patch]: Release 0.1.23 (#25804)
openai[patch]: fix get_num_tokens for function calls (#25785)
all: Improve make lint command (#25344)
openai:compatible with other llm usage meta data (#24500)
core[patch]: add additionalProperties recursively to oai function if strict (#25169)
docs: fix openai api ref (#25639)
langchain-ollama==0.1.2
Changes since langchain-ollama==0.1.1
ollama: bump core version (#25826)
partners/ollama: release 0.1.2 (#25817)
all: Improve make lint command (#25344)
add embeddings integration tests (#25508)
core, partners: implement standard tracing params for LLMs (#25410)
ollama[patch]: Update extra in ollama package (#25383)
ollama[patch]: Update API Reference for ollama embeddings (#25315)
ollama: allow base_url, headers, and auth to be passed (#25078)
patch[Partners] Unified fix of incorrect variable declarations in all check_imports (#25014)
langchain-core==0.2.36
langchain-community==0.2.14
langchain-community==0.2.13
Changes since langchain-community==0.2.12
community[patch]: Release 0.2.13 (#25806)
community[patch]: Minor Improvement of extract hyperlinks tool output (#25728)
docs minor fix (#25794)
Add mmr to neo4j vector (#25765)
community: Azure Search Vector Store is missing Access Token Authentication (#24330)
Improvement[Community] Improve methods in IMessageChatLoader (#25746)
Fix MoonshotChat instantiate with alias (#25755)
Update endpoint for AzureMLEndpointApiType class. (#25725)
[Community][minor]: Added langchain_version while calling discover API (#24428)
[Community][minor]: Updating source path, and file path for SharePoint loader in PebbloSafeLoader (#25592)
community[minor]: [Pebblo] Fix URL construction in newer Python versions (#25747)
community: [PebbloRetrievalQA] Implemented Async support for prompt APIs (#25748)
all: Improve make lint command (#25344)
community: remove integration test deps (#24460)
Added support for Nebula Chat model (#21925)
: tidb vector support vector index (#19984)
community:yuan2[patch]: standardize init args (#21462)
community: remove unused verify_ssl kwarg from aiohttp request (#25707)
docs: Add docstrings for CassandraChatMessageHistory class and package namespace function. (#24222)
community: NeptuneGraph fix (#23281)
community: Opensearch - added score function for similarity_score_threshold (#23928)
patch[Community] Optimize methods in several ChatLoaders (#24806)
Use extend method for embeddings concatenation in mlflow_gateway (#14358)
partners: Use simsimd types (#25299)
community: add supported blockchains to Blockchain Document Loader (#25428)
community: allow chroma DB delete() to use "where" argument (#19826)
community[minor]: Fix missing 'keep_newlines' parameter forward-pass to 'process_pages' function in confluence loader (#20086) (#20087)
community[patch]: Fix Hybrid Search for non-Databricks managed embeddings (#25590)
community[patch]: Updating default PPLX model to supported llama-3.1 model. (#25643)
community[patch]: Dynamodb history messages key (#25658)
community: fix HuggingFacePipeline pipeline_kwargs (#19920)
"community: Fix GithubFileLoader source code", "docs: Fix GithubFileLoader code sample" (#19943)
langchain_community.graphs: Neo4JGraph: prop min_size might be None (#23944)
Rm DeepInfra Breakpoint Comment (#25206)
community: add metadata filter to CassandraGraphVectorStore (#25663)
community: Give more context on DeepInfra 500 errors (#25671)
community: Refactor PebbloRetrievalQA (#25583)
community: Refactor PebbloSafeLoader (#25582)
docs: Add ChatZhipuAI tool calling and structured output docstring (#25669)
community: add retry for session expired exception in neo4j (#25660)
[Community][minor]: Updating metadata with full_path in SharePoint loader (#25593)
Update azuresearch.py (#25577)
core,community: add beta decorator to missed GraphVectorStore extensions (#25562)
community[patch]: Add ToolMessage for ChatZhipuAI (#25547)
[Community]: Fix - Open AI Whisper client.audio.transcriptions returning Text Object which raises error (#25271)
docs: Fix QianfanLLMEndpoint and Tongyi input text (#25529)
Improvement[Community] Improve api doc for SingleFileFacebookMessengerChatLoader (#25536)
core, partners: implement standard tracing params for LLMs (#25410)
docs: fix mimetype parser docstring (#25463)
langchain[patch]: deprecate various chains (#25310)
community[patch]: update connection string in azure cosmos integration test (#25438)
[docs]: doc loader changes (#25417)
fix: typo where github should be gitlab (#25397)
docs: udpated api reference (#25172)
multiple: update removal targets (#25361)
docs: improve document loaders index (#25365)
[docs]: standardize doc loader doc strings (#25325)
[docs]: standardize tool docstrings (#25351)
community: fix AzureSearch vectorstore asyncronous methods (#24921)
community: add args_schema to SearxSearchResults tool (#25350)
Standardize SparkLLM (#25239)
qianfan generate/agenerate with usage_metadata (#25332)