Changes since langchain-qdrant==0.2.0

release(qdrant): 0.2.1 (#32889)
fix(qdrant): allow as_retriever to work without embeddings in SPARSE mode (#32757)
chore: remove ruff target-version (#32880)
chore: adress pytest-asyncio deprecation warnings + other nits (#32696)
feat(qdrant): ruff fixes and rules (#32500)
fix: formatting issues in docstrings (#32265)
fix(docs): capitalization, codeblock formatting, and hyperlinks, note blocks (#32235)
ruff: more rules across the board & fixes (#31898)
fix: automatically fix issues with ruff (#31897)
langchain-qdrant[patch]: Add ruff bandit rules to linter (#31815)
packaging: remove Python upper bound for langchain and co libs (#31025)
partners[lint]: run pyupgrade to get code in line with 3.9 standards (#30781)
langchain_qdrant: fix showing the missing sparse vector name (#30701)
multiple: fix uv path deps (#29790)
qdrant: Add similarity_search_with_score_by_vector() function to the QdrantVectorStore (#29641)
infra: add UV_FROZEN to makefiles (#29642)
infra: migrate to uv (#29566)
multiple: disable socket for unit tests (#29080)
qdrant[patch]: run python 3.13 in CI (#28394)
Improvement[Partner] Improve qdrant vector store (#27251)

Changes since langchain-openai==0.3.32

release(openai): 0.3.33 (#32887)
chore: remove ruff target-version (#32880)
fix(openai): structured output (#32551)
fix(openai): handle AIMessages without response_id in _get_last_messages (#32824)
fix(openai): Fix Azure OpenAI Responses API model field issue (#32649)
chore(openai): apply formatting changes to AzureChatOpenAI (#32848)
feat(core): add web_search in OpenAI tools list (#32738)
docs: standardize OllamaLLM and BaseOpenAI docstrings (#32758)
chore: adress pytest-asyncio deprecation warnings + other nits (#32696)

Changes since langchain-core==0.3.75

release(core): 0.3.76 (#32877)
fix(core): honor enabled=false in nested tracing (#31986)
fix(core): ensure InjectedToolCallId always overrides LLM-generated values (#32766)
chore(docs): update package READMEs (#32869)
chore(core): cleanup pyproject.toml (#32865)
refactor(core): improve beta decorator (#32505)
feat(core): add id field to Document passed to filter for InMemoryVectorStore similarity search (#32688)
feat(core): support AWS Bedrock document content blocks in msg_content_output (#32799)
feat(core): support for adding PromptTemplates with formats other than f-string (#32253)
chore(core): add ruff rule PLC0415 (#32351)
chore(core): fix some pydocs to use google-style (#32764)
chore(core): fix some ruff preview rules (#32785)
chore(core): fix some docstrings (from DOC preview rule) (#32833)
refactor(core): use pytest style in TestGetBufferString (#32786)
chore(core): enable ruff docstring-code-format (#32834)
chore(core): fix some mypy warn_unreachable issues (#32560)
feat(core): add web_search in OpenAI tools list (#32738)
chore: adress pytest-asyncio deprecation warnings + other nits (#32696)

Changes since langchain-groq==0.3.7

release(groq): 0.3.8 (#32870)
feat(groq): add support for json_schema (#32396)
chore: adress pytest-asyncio deprecation warnings + other nits (#32696)
feat: port various nit changes from wip-v0.4 (#32506)

Initial release

feat(langchain): middleware support in create_agent (#32828)

Initial release

release(langchain): v1.0.0a3 (#32791)
chore(langchain): rename create_react_agent -> create_agent (#32789)
chore(langchain): remove upper bound at v1 for core (#32737)
chore(langchain): also bump text splitters (#32722)
chore(langchain): use latest core (#32720)
chore(langchain): revert back to static versioning for now (#32719)
release(langchain): v1.0.0a1 (#32718)
feat(langchain): revamped create_react_agent (#32705)
chore(langchain): remove untested chains for first alpha (#32710)
chore(langchain): drop Python 3.9 to prep for v1 (#32704)
chore: adress pytest-asyncio deprecation warnings + other nits (#32696)
chore: update references to use the latest version of Claude-3.5 Sonnet (#32594)
docs: update outdated README.md content (#32540)
fix(tests): add anthropic_proxy to configurable test parameters (for v1)
chore: formatting across codebase (#32466)
feat(openai): minimal and verbosity (#32455)
feat(langchain): add stuff and map reduce chains (#32333)
fix: use new Google model names in examples (#32288)
fix: update bar_model to use the correct model version claude-3-7-sonnet-20250219 (#32284)
refactor: remove references to unsupported model claude-3-sonnet-20240229 (#32281)
fix: formatting issues in docstrings (#32265)
feat(docs): improve devx, fix Makefile targets (#32237)
chore(langchain_v1): clean anything uncertain (#32228)
feat(langchain): v1 scaffolding (#32166)

When interacting with the Responses API, langchain-openai now defaults to storing response items in message content. This behavior was previously opt-in by specifying output_version="responses/v1" when instantiating ChatOpenAI. This was done to resolve BadRequestErrors that can arise in some multi-turn contexts.

To restore previous behavior, set the LC_OUTPUT_VERSION environment variable to v0, or specify output_version="v0" when instantiating ChatOpenAI:

os.environ["LC_OUTPUT_VERSION"] = "v0"

# or

from langchain_openai import ChatOpenAI

llm = ChatOpenA(model="...", output_version="v0")

Initial release

release(core): 1.0.0a2 (#32792)
fix: ci
fix some imports
rfc: remove unused TypeGuards
chore: move convert_to_openai_data_block and convert_to_openai_image_block from content.py to openai block translators
chore: move _convert_openai_format_to_data_block from langchain_v0 to openai
fix(core): typo in content.py
release(core): 1.0.0a1 (#32715)
release(core): 1.0.0dev0 (#32713)
chore: (v1) drop support for python 3.9 (#32712)
feat: standard content, IDs, translators, & normalization (#32569)
Merge branch 'master' into wip-v1.0
chore: adress pytest-asyncio deprecation warnings + other nits (#32696)
release(core): 0.3.75 (#32693)
feat(core): trace response body on error (#32653)
chore(core): add note about backward compatibility for tool_calls in additional_kwargs in JsonOutputKeyToolsParser
docs(docs): fixed typos in documentations (#32661)
Merge branch 'master' into wip-v1.0
fix(core): (v1) update BaseChatModel return type to AIMessage (#32626)
chore(core): add mypy pydantic plugin (#32604)
refactor(core): Use duck typing for _StreamingCallbackHandler (#32535)
Merge branch 'master' into wip-v1.0
fix: use .get() on image URL in ImagePromptValue.to_string()
fix(core): Support no-args tools by defaulting args to empty dict (#32530)
docs: update outdated README.md content (#32540)
chore(core): bump lock
feat: port various nit changes from wip-v0.4 (#32506)
fix(core): fix beta decorator for properties (#32497)
chore: formatting across codebase (#32466)
release(core): 0.3.74 (#32454)
feat(openai): custom tools (#32449)
release(core): 0.3.73 (#32446)
feat(core): zero-out token costs for cache hits (#32437)
fix(core): fix index checking when merging lists (#32431)
fix(core): revert "fix: tool call streaming bug with inconsistent indices from Qwen3" (#32307)
fix: lint
fix: tool call streaming bug with inconsistent indices from Qwen3 (#32160)
fix(core): resolve cache validation error by safely converting Generation to ChatGeneration objects (#32156)
chore(core): add test for nested pydantic fields in schemas (#32285)
refactor: remove references to unsupported model claude-3-sonnet-20240229 (#32281)
fix(core): track within-batch deduplication in indexing num_skipped count (#32273)
fix: formatting issues in docstrings (#32265)
feat(docs): improve devx, fix Makefile targets (#32237)
fix: LLM mimicking Unicode responses due to forced Unicode conversion of non-ASCII characters. (#32222)
fix(docs): capitalization, codeblock formatting, and hyperlinks, note blocks (#32235)
release(core): 0.3.72 (#32214)
fix(core): Dereference Refs for pydantic schema fails in tool schema generation (#32203)
docs: formatting cleanup (#32188)
release(core): 0.3.71 (#32186)
fix(core): exclude api_key from tracing metadata (#32184)
fix(core): add type key when tracing in some cases (#31825)
feat(core): add sanitize_for_postgres utility to fix PostgreSQL NUL byte DataError (#32157)
fix(core): fix parse_resultin case of self.first_tool_only with multiple keys matching for JsonOutputKeyToolsParser (#32106)
release(core): 0.3.70 (#32144)
docs(core): move incorrect arg limitation in rate limiter's docstring (#32118)
fix(core): fix docstrings and add sleep to FakeListChatModel._call (#32108)
fix(core): better error message for empty var names (#32073)
fix(core): implement sleep delay in FakeMessagesListChatModel _generate (#32014)
fix(core): JSON Schema reference resolution for list indices (#32088)
feat(core): update _import_utils.py to not mask the thrown exception (#32071)
fix: fixing missing Docstring Bug if no Docstring is provided in BaseModel class (#31608)
fix: issue a warning if np.nan or np.inf are in _cosine_similarity argument Matrices (#31532)
release(core): release 0.3.69 (#32056)
feat(core): add an option to make deserialization more permissive (#32054)
chore(core): unpin packaging dependency (#32032)
docs: add deprecation notice for PipelinePromptTemplate (#31999)
fix(core): correct return type hints in BaseChatPromptTemplate (#32009)
feat(core): add ruff rules D to tests except D1 (#32000)
fix[core]: added error message if the query vector or embedding contains NaN values (#31822)
ruff: restore stacklevels, disable autofixing (#31919)
core: docs: clarify where the kwargs in on_tool_start and on_tool_end go (#31909)
ruff: more rules across the board & fixes (#31898)
ollama: thinking, tool streaming, docs, tests (#31772)
fix: fix file open with encoding in chat_history.py (#31884)
core: Cleanup pyproject (#31857)
fix: remove unused type ignore from three_values fixture in TestAsyncInMemoryStore (#31895)
core: Ruff preview fixes (#31877)
core: make ruff rule PLW1510 unfixable (#31868)
core: Bump ruff version to 0.12 (#31846)
core[patch]: Int Combine when Merging Dicts (#31572)
core: release 0.3.68 (#31848)
core: Use parametric tests in test_openai_tools (#31839)
core[path]: Use context manager for FileCallbackHandler (#31813)
core: release 0.3.67 (#31791)
core: expose tool message recognized block types (#31787)
docs: enhance docstring for disable_streaming parameter in BaseChatModel (#31759)
fix: update import paths for ChatOllama to use langchain_ollama instead of community (#31721)
core[patch]: Add additional hashing options to indexing API, warn on SHA-1 (#31649)
standard-tests: refactoring and fixes (#31703)
core: Use parametrized test in test_correct_get_tracer_project (#31513)
docs: document Anthropic cache TTL count details (#31708)
core[patch]: Improve RunnableWithMessageHistory init arg types (#31639)
core: fix get_buffer_string output for structured message content (#31600)
core[patch]: check before removing tags (#31691)
core: release 0.3.66 (#31690)
core[patch]: Returning always 2D Array for _cosine_similarity (#31528)
core[patch]: Add doc-strings to tools/base.py (#31684)
core: Cleanup Pydantic models and handle deprecation warnings (#30799)
core and langchain: Remove upper bound restriction langsmith dependency (#31629)
bump langsmith to allow 0.4 (#31594)
core: In LangChainTracer, send only the first token event (#31591)
core: release 0.3.65 (#31557)
core(fix): revert set_text optimization (#31555)
core: use run tree post/patch (#31500)
core[patch]: Correct type casting of annotations in _infer_arg_descriptions (#31181)
core[patch]: revert change to stream type hint (#31501)
core: Add mypy strict-equality rule (#31286)
fix: module 'defusedxml' has no attribute 'ElementTree' (#31429) (#31431)
core: release 0.3.63 (#31419)
langchain-core: Add image_generation tool to list of known openai tools (#31396)
standard-tests: add benchmarks (#31302)
core: release 0.3.62 (#31376)
anthropic: support for code execution, MCP connector, files API features (#31340)
core: release 0.3.61 (#31317)
openai[patch]: support built-in code interpreter and remote MCP tools (#31304)
core: Improve Runnable __or__ method typing annotations (#31273)
core: support Union type args in strict mode of OpenAI function calling / structured output (#30971)
core: Add ruff rules A (builtins shadowing) (#29312)
langchain[patch]: Allow async indexing code to work for vectorstores that only defined sync delete (#30869)
core: Add ruff rules RUF (#29353)
core: release 0.3.60 (#31249)
core: Add ruff rules SLF (#30666)
perf[core]: remove unnecessary model validators (#31238)
packaging[core]: bump min pydantic version (#31239)
langchain-core[patch]: Incremental record manager deletion should be batched (#31206)
perf[core]: remove generations summation from hot loop (#31231)
perf[core]: remove costly async helpers for non-end event handlers (#31230)
core: Fix some private member accesses (#30912)
core: Fix issue 31035 alias fields in base tool langchain core (#31112)
core: release 0.3.59 (#31150)
fix(core): Permit OpenAI style blocks to be passed into convert_to_openai_messages (#31140)
core: release 0.3.58 (#31099)
core, openai[patch]: prefer provider-assigned IDs when aggregating message chunks (#31080)
0.3.57 (#31095)
Enable run mutation (#31090)
packaging: remove Python upper bound for langchain and co libs (#31025)
core: release 0.3.56 (#31000)
core: release 0.3.56rc1 (#30998)
core[patch]: fix edge cases for _is_openai_data_block (#30997)
feat(core): Autogenerate filenames for when converting file content blocks to OpenAI format (#30984)
core, standard-tests: support PDF and audio input in Chat Completions format (#30979)
core[patch]: update dict prompt template (#30967)
core, openai: support standard multi-modal blocks in convert_to_openai_messages (#30968)
core[patch]: release 0.3.55 (#30952)
core: Make Graph.Node.data optional (#30943)
core: Improve OutputParser error messaging when model output is truncated (max_tokens) (#30936)
core[patch]: de-beta usage callback (#30928)
core: release 0.3.54 (#30911)
core[patch]: add retries and better messages to draw_mermaid_png (#30881)
core[patch]: Raise AttributeError (instead of ModuleNotFoundError) in custom __getattr__ (#30905)
core: release 0.3.53 (#30901)
multiple: permit optional fields on multimodal content blocks (#30887)
core: Removing unnecessary pydantic core schema rebuilds (#30848)
core[fix]: Fix __dir__ in __init__.py for output_parsers module (#30856)
core: Remove some noqa (#30855)
core[fix]: remove load from dynamic imports dict (#30849)
core: release 0.3.52 (#30850)
Fix from langchain_core.load.load import load import (#30843)
core[patch]: dict chat prompt template support (#25674)
multiple: multi-modal content blocks (#30746)
Tinkering with CodSpeed (#30824)
Consistent docstring indentation (#30834)
core[lint]: fix issue with unused ignore in __init__.py files (#30825)
core[performance]: use custom __getattr__ in __init__.py files for lazy imports (#30769)
core: Improve mypy config (#30737)
core: Add ruff rules FBT001 and FBT002 (#30695)
core[patch]: Share executor for async callbacks run in sync context (#30779)
core[lint]: Use 3.9 formatting for docs and tests (#30780)
core[lint]: Fix typing in `...

Changes since langchain-text-splitters==0.3.10

release(text-splitters): 0.3.11 (#32770)
docs(text-splitters): fix some docstrings (#32767)

Changes since langchain-cli==0.0.36

release(cli): 0.0.37 (#32760)
chore(cli): add mypy strict checking (#32386)
chore(cli): add ruff rules ANN401 and D1 (#32576)
chore: adress pytest-asyncio deprecation warnings + other nits (#32696)
chore(cli): select ALL rules with exclusions (#31936)
chore(cli): add ruff rules D1 (#32350)
chore: formatting across codebase (#32466)
fix(docs): local API reference documentation build (#32271)
fix: formatting issues in docstrings (#32265)
feat(docs): improve devx, fix Makefile targets (#32237)
fix: LLM mimicking Unicode responses due to forced Unicode conversion of non-ASCII characters. (#32222)
fix(cli): exception to prevent swallowing unexpected errors (#31983)
fix(cli): handle exception in remove() (#31982)
ruff: add bugbear across packages (#31917)
cli: ensure the connection always get closed in github.py (#31914)
huggingface[patch]: ruff fixes and rules (#31912)
ruff: more rules across the board & fixes (#31898)
fix: automatically fix issues with ruff (#31897)
cli: Ruff autofixes (#31863)
cli: fix dockerfile incorrect copy (#31883)
cli: Bump ruff version to 0.12 (#31864)
cli: Add ruff rule UP (pyupgrade) (#31843)
docs: fix grammar in multiple docs (#31375)
ci: fix typo in doc-string (#31284)
packaging: remove Python upper bound for langchain and co libs (#31025)
docs: streamline LangSmith teasing (#30302)
standard-tests[patch]: require model_name in response_metadata if returns_usage_metadata (#30497)
cli: use pytest-watcher (#30465)
cli: update integration doc template for tools (#30188)