LangChain Release Notes

Changes since langchain-perplexity==1.4.0

• release(perplexity): 1.4.1 (#39826)
• fix(perplexity): include type="message" on Responses input items (#39774)
• fix(perplexity): preserve caller extra_body (#39203)
• chore: bump pillow from 12.2.0 to 12.3.0 in /libs/partners/perplexity (#38991)
• chore(deps): refresh lockfiles (#38746)
• fix(perplexity): use supported Responses API model (#38337)
• chore: bump vcrpy from 8.1.1 to 8.2.1 in /libs/partners/perplexity (#38290)
• chore: bump langsmith from 0.8.5 to 0.8.18 in /libs/partners/perplexity (#38289)
• docs: refresh README installation and resources (#38119)
• release(core): 1.4.7 (#38111)
• fix(core,partners): rename package version trace metadata (#38110)
• style(core,langchain,langchain-classic,partners): replace double backticks in docstrings (#38095)
• release(core): 1.4.6 (#38061)
• feat(core,partners): add package version tracking to tracing metadata (#35295)
• chore(infra): bump mypy to 2.1 and unify type-check config across the monorepo (#36470)
• feat(standard-tests): validate tool call chunks during streaming (#34707)
• chore(partners): bump locks (#38052)

Changes since langchain-fireworks==1.5.2

release(fireworks): 1.6.0 (#39810)

feat(fireworks): add document reranking (#39732)

fix(fireworks): filter invalid tool calls from v1 content (#39805)

feat(core): add standard model exception types (#39538)

chore(model-profiles): refresh model profile data (#39646)

fix(fireworks): avoid repeated service_tier [closes #39619] (#39620)

chore: bump langsmith from 0.10.6 to 0.10.16 in /libs/partners/fireworks (#39279)

chore: bump aiohttp from 3.14.1 to 3.14.3 in /libs/partners/fireworks (#39236)

chore(model-profiles): refresh model profile data (#39166)

Changes since langchain-core==1.5.6

release(core): 1.6.0 (#39760)

fix(core): resolve postponed annotations in StructuredTool._injected_args_keys (#39602)

feat(core): add standard model exception types (#39538)

fix(core): allow deserializing RunnablePick (#39753)

fix(core): make convert_to_openai_function handle callables and non-dict mappings (#39750)

fix(core): make subprocess and temporary file tests portable on Windows (#39664)

fix(core): fail fast when tool schemas can't resolve forward refs during serialization (#39570)

test(core): avoid version-dependent runnable snapshots (#39705)

fix(core): require all nested properties for strict tool schemas (#39306)

fix(core): remove stale sync-stream xfail [closes #39720] (#39723)

perf(core): Lazily import transformers (#38037)

fix(core): accept non-dict Mapping values in mustache templates (#39680)

docs(core): clarify Runnable pipe coercion [closes #39075] (#39707)

fix(core): finalize chain-group runs on BaseException (#39699)

Changes since sdk==0.4.2

• release(sdk-py): 0.4.3 (#8657)
• feat(sdk-py): add decrypt replacement result (#8598)
• release(langgraph): 1.2.11 (#8595)
• chore(deps): bump the minor-and-patch group across 1 directory with 5 updates (#8532)
• release(checkpoint): 4.2.0 (#8563)
• chore: enforce PLC0415 in tests for the remaining packages (#8547)
• release(langgraph): 1.2.10 (#8462)
• chore(deps): bump websockets from 15.0.1 to 16.0 in /libs/sdk-py in the major group (#8253)
• fix(sdk-py): support clearing cron end_time via update(end_time=None) (#8334)
• release(langgraph): 1.2.9 (#8316)
• release(langgraph): 1.2.8 (#8292)
• chore(deps): bump websockets from 15.0.1 to 16.0 in /libs/langgraph in the major group (#8256)
• chore(deps): bump the minor-and-patch group in /libs/sdk-py with 9 updates (#8252)
• release(langgraph): 1.2.7 (#8223)
• chore(deps-dev): bump starlette from 1.0.1 to 1.3.1 in /libs/sdk-py (#8104)
• chore(deps): bump langsmith from 0.8.0 to 0.8.18 in /libs/sdk-py (#8174)
• release(langgraph): 1.2.6 (#8139)
• docs: standardize package README.md structure (#8064)
• release(langgraph): 1.2.5 (#8062)
• fix(langgraph): merge lc_versions config metadata (#8052)
• chore(deps-dev): bump starlette from 1.0.0 to 1.0.1 in /libs/sdk-py (#8006)
• chore: migrate Python type checking to ty (#8002)
• release(langgraph): 1.2.4 (#7991)
• test(sdk-py): add factory-graph integration test exercising the server factory path (#7978)
• release(langgraph): 1.2.3 (#7945)

Initial release
release(openai): 1.5.2a1 (#39709)
feat(openai): extract gateway metadata from response headers when available (#39706)
chore(openai): update snapshots (#39657)
fix(openai): support o-series models in get_num_tokens_from_messages (#38710)
release(openai): 1.5.1 (#39653)
fix(openai): preserve streamed encrypted reasoning (#39635)
chore(infra): support langsmith gateway in CI (#39651)
release(openai): 1.5.0 (#39629)
feat(openai): support openai 3.0 SDK (#39613)
chore(partners): bump langgraph floor in openai and huggingface lockfiles (#39617)
release(openai): 1.4.3 (#39485)
fix(openai): filter invalid tool calls from content (#39366)
chore(openai): update guidance for responses API for OpenAI-compatible providers (#39327)
chore(openai): update docstring for include_response_headers (#39326)
release(openai): 1.4.2 (#39322)
fix(openai): handle ContextWindowExceededError (#39300)
chore: bump the minor-and-patch group across 3 directories with 7 updates (#39187)
fix(openai): filter langchain-generated content block IDs (#39209)
fix(openai): preserve Responses text options (#39204)
fix(openai): redact MCP authorization (#39155)
chore(model-profiles): refresh model profile data (#39050)
release(openai): 1.4.1 (#39045)
feat(anthropic,fireworks,openai): support langsmith gateway through env var (#38742)
fix(openai): correct gpt-5.3-chat-latest profile (#39009)
release(openai): 1.4.0 (#38983)
chore: bump pillow from 12.2.0 to 12.3.0 in /libs/partners/openai (#38999)
feat(core): add reasoning_effort as a standard chat model parameter (#38887)
chore(model-profiles): refresh model profile data (#38797)
release(openai): 1.3.5 (#38785)
feat(openai): support explicit prompt caching (#38762)
chore(model-profiles): refresh model profile data (#38774)
release(openai): 1.3.4 (#38731)
fix(openai): suppress Pydantic serializer warning on structured output parsed field (#37727)
test(openai): skip Codex VCR tests before cassette setup (#38690)
chore: bump the minor-and-patch group across 3 directories with 11 updates (#38587)
chore: bump langgraph-checkpoint from 4.1.0 to 4.1.1 in /libs/partners/openai (#38476)
fix(core): use asyncio.get_running_loop() in async contexts (#38157)
test(openai): clarify async API key sync failure trace (#38379)
release(openai): 1.3.3 (#38375)
fix(openai): drop response item ids when store is false (#38372)
fix(langchain,openai): only set strict=True on tools for OpenAI-compatible models in ProviderStrategy (#38370)
test(openai): clarify expected strict schema error (#38338)
fix(openai): drop stop from Responses API payload (#38336)
chore: bump langsmith from 0.8.5 to 0.8.18 in /libs/partners/openai (#38293)
chore: bump vcrpy from 8.1.1 to 8.2.1 in /libs/partners/openai (#38294)
chore(model-profiles): refresh model profile data (#38274)
test(openai): vcr embedding raw equivalence tests (#38199)
release(openai): 1.3.2 (#38130)
fix(openai): build Codex async headers off the event loop in _agenerate (#38129)
fix(openai): avoid sync token reads in Codex streaming (#38128)
hotfix(openai): skip Codex live integration tests in CI (#38124)
hotfix(openai): switch version (#38123)
refactor(openai): mark Codex OAuth classes private (#38122)
release(openai): 1.4.0 (#38120)
feat(openai): add ChatGPT OAuth-backed ChatOpenAICodex chat model (#37569)
docs: refresh README installation and resources (#38119)
test(core,langchain): update tests for explicit deserialization allowlists (#38118)
release(core): 1.4.7 (#38111)
fix(core,partners): rename package version trace metadata (#38110)
style(core,langchain,langchain-classic,partners): replace double backticks in docstrings (#38095)
test(openai): use gpt-4o for image token counting (#38089)
release(core): 1.4.6 (#38061)
feat(core,partners): add package version tracking to tracing metadata (#35295)
fix(core,openai): normalize v1 streamed tool calls (#35983)
chore(infra): bump mypy to 2.1 and unify type-check config across the monorepo (#36470)
feat(standard-tests): validate tool call chunks during streaming (#34707)
fix(langchain): tighten structured output model fallbacks (#38042)
hotfix(openai): min core dep (#37990)
release(openai): 1.3.0 (#37989)
feat(openai): support apply_patch built-in tool (#37157)
chore(model-profiles): refresh model profile data (#37973)
test(langchain,partners): disable pytest-benchmark under xdist to silence PytestBenchmarkWarning (#37901)
chore(model-profiles): refresh model profile data (#37895)
release(openai): 1.2.2 (#37617)
chore(infra): bump langchain-tests floor to 1.1.9 (#37610)
test(openai): unbreak audio chat and Azure embedding integration tests (#37589)
fix(openai): guard httpx finalizers (#37570)
chore: bump langsmith from 0.8.4 to 0.8.5 in /libs/partners/openai (#37549)
chore: bump idna from 3.11 to 3.15 in /libs/partners/openai (#37548)
ci(infra): harden Dependabot version-bound preservation (#37510)
test(standard-tests): assert ls_model_name honors per-call model override (#37504)
fix(openai): source LLM context size from model profiles (#37489)
chore(core,langchain,openai): refresh stale OpenAI model references (#37487)
fix(openai): broaden condition for ContextOverflowError to accommodate other providers (#37457)
docs(openai): document base_url env var fallback chain (#37436)
chore: bump langsmith from 0.8.0 to 0.8.4 in /libs/partners/openai (#37416)
chore: bump langsmith from 0.7.31 to 0.8.0 in /libs/partners/openai (#37398)
chore(infra): merge v1.4 into master (#37350)
chore: bump urllib3 from 2.6.3 to 2.7.0 in /libs/partners/openai (#37330)
chore: bump langchain-core from 1.3.2 to 1.3.3 in /libs/partners/openai (#37266)
chore(docs): update x handle references (#37081)
chore(model-profiles): refresh model profile data (#37074)
chore(docs): update comment for chatopenai (#37034)
chore(model-profiles): refresh model profile data (#37015)
hotfix: bump min core versions (#36996)
release(openai): 1.2.1 (#36995)
fix(openai): add gpt-5.5 pro to Responses API check (#36994)
feat(core): add content-block-centric streaming (v2) (#36834)
chore(model-profiles): refresh model profile data (#36982)
release(openai): 1.2.0 (#36961)
feat(openai): prevent silent streaming hangs in ChatOpenAI (#36949)
hotfix(ci): remove nobenchmark flag (#36959)
chore(partners): standardize integration test invocation (#36958)
release(openai): 1.1.16 (#36927)
fix(openai): tolerate prompt_cache_retention drift in streaming (#36925)
release(openai): 1.1.15 (#36901)
fix(openai): accommodate dict response items in streaming (#36899)
fix(openai): infer azure chat profiles from model name (#36858)
chore(model-profiles): refresh model profile data (#36864)
release(openai): 1.1.14 (#36820)
fix(openai): use SSRF-safe transport for image token counting (#36819)
chore(deps): bump pytest to 9.0.3 (#36801)
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/partners/openai (#36795)
chore: bump pillow from 12.1.1 to 12.2.0 in /libs/partners/openai (#36777)
release(openai): 1.1.13 (#36729)
fix(openai): handle content blocks without type key in responses api conversion (#36725)
chore(model-profiles): refresh model profile data (#36539)
chore(openai): fix broken vcr cassette playback and add ci guard (#36502)
fix(openai,groq,openrouter): use is-not-None checks in usage metadata token extraction (#36500)
fix(core): fixed typos in the documentation (#36459)
chore(model-profiles): refresh model profile data (#36455)
feat(core): impute placeholder filenames for OpenAI file inputs (#36433)
chore: pygments>=2.20.0 across all packages (CVE-2026-4539) (#36385)
chore(model-profiles): refresh model profile data (#36368)
fix(openai): update computer call test (#36352)
fix(openai): let user-provided User-Agent override the Azure default (#35523)
chore: bump requests from 2.32.5 to 2.33.0 in /libs/partners/openai (#36248)
fix(openai): bump min core version (#36180)
release(openai): 1.1.12 (#36178)
fix(core,model-profiles): add missing ModelProfile fields, warn on schema drift (#36129)
fix(openai): support phase parameter (#36161)
fix(openai): preserve namespace field in streaming function_call chunks (#36108)
ci: suppress pytest streaming output in CI (#36092)
ci: avoid unnecessary dep installs in lint targets (#36046)
chore(model-profiles): refresh model profile data (#36039)
chore: bump orjson from 3.11.5 to 3.11.6 in /libs/partners/openai (#35860)
fix(openai): add type: message to Responses API input items (#35693)
perf(.github): set a timeout on get min versions HTTP calls (#35851)
feat(model-profiles): new fields + Makefile target (#35788)
fix(openai): close PIL Image handles in token counting to prevent fd leak (#35742)
fix(openai): typo (#35763)
chore(model-profiles): refresh model profile data (#35754)
fix(openai): bump min core version (#35705)
release(openai): 1.1.11 (#35703)
fix(openai): update responses API model detection for pro and codex models (#35594)
feat(openai): support tool search (#35582)
chore: bump langgraph from 1.0.8 to 1.0.10rc1 in /libs/partners/openai (#35612)
chore(model-profiles): refresh model profile data (#35593)
fix(openai): avoid PydanticSerializationUnexpectedValue for structured output (#35543)
feat(openrouter): add streaming token usage support (#35559)
fix: compaction typo (#35467)
fix(openai): add test for CSV and accommodate breaking changes in file url inputs (#35454)
chore: bump langgraph-checkpoint from 3.0.0 to 4.0.0 in /libs/partners/openai (#35448)
fix(model-profiles): sort generated profiles by model ID for stable diffs (#35344)
fix(openai): accept valid responses that are falsy at runtime (#35307)
release(openai): 1.1.10 (#35292)
feat(openai): support automatic server-side compaction (#35212)
fix(openai): add model property (#35284)
fix(nomic,openai,perplexity): update pillow version to >= 12.1.1, <13.0.0 (#35254)
docs(openai): more nits (#35277)
docs(openai): clarify reasoning config for openai-compatible endpoints (#35202)
fix(openai): gpt-5.2-pro Model Profile structured_output key fixed (#35216)
chore(openai): extend model_token_mapping till gpt-5.2 for modelname_to_contextsize (#35214)
fix(openai): enhance error message for non-OpenAI embedding providers (#35252)
fix(openai): sanitize chat completions text content blocks (#35217)
fix(openai): improve error message for null choices in OpenAI-compatible APIs (#35236)
feat(openrouter): add langchain-openrouter provider package (#35211)
chore: bump pillow from 11.3.0 to 12.1.1 in /libs/partners/openai (#35177)
chore(deps): bump langsmith from 0.4.37 to 0.6.3 in /libs/partners/openai (#35151)
release(openai): 1.1.9 (#35145)
fix(openai): sanitize urls when counting tokens in images (#35143)
feat(core): add ContextOverflowError, raise in anthropic and openai (#35099)
feat(model-profiles): add text_inputs and text_outputs (#35084)
release(openai): 1.1.8 (#35097)
fix(openai): detect codex models for responses api preference (#35058)
chore: add make type target (#35015)
revert: "chore: add typing target in Makefile" (#35013)
chore: add typing target in Makefile (#35012)
chore: enrich pyproject.toml files (#34980)
chore(deps): bump the uv group across 20 directories with 3 updates (#34941)
chore(openai): update model profiles (#34960)
chore: upgrade urllib3 to 2.6.3 (#34940)
chore(deps): bump the uv group across 8 directories with 6 updates (#34773)
chore: update twitter URLs (#34736)
chore: ban relative imports on all packages (#34691)
chore(openai): delete outdated test (#34682)
release(openai): 1.1.7 (#34640)
fix(openai): raise proper exception OpenAIRefusalError on structured output refusal (#34619)
fix: handle empty assistant content in Responses API (#34272) (#34296)
fix(core): serialization patch (#34455)
fix(openai): handle function_call content in token counting (#34379)
fix(openai): filter function_call blocks in token counting (#34396)
release(openai): 1.1.6: update max input tokens for gpt-5 series (#34419)
release(openai): 1.1.5 (#34409)
fix(openai): rely on langchain-core for setting chunk_position (#34404)
release(openai): 1.1.4 (#34395)
revert(openai): switch model from nano to mini when using flex (#34394)
fix(openai): Correct hyperlinks in documentation of function with_structured_output (#34385)
test(openai): switch model from nano to mini when using flex (#34336)
revert: make integration tests output verbose (#34329)
chore(infra,openai): make integration tests output verbose (#34327)
release(openai): 1.1.3 (#34325)
fix(openai): correct image resize aspect ratio caps (#34192)
refactor(standard-tests): improve VCR config (#33968)
feat(openai): Use responses API if model is gpt-5.2-pro (#34306)
fix(openai): handle missing 'text' key in responses API content blocks (#34198)
release(openai): 1.1.2 (#34302)
fix(openai): Allow temperature when reasoning is set to the string 'none' (#34298)
feat(langchain,openai): add strict flag to ProviderStrategy structured output (#34149)
docs(fireworks,groq,huggingface,mistralai,ollama,openai): x-ref convert_to_openai_tool (#34276)
fix(openai): add missing tools param to ChatOpenAI with_structured_output (#34075)
release(openai): 1.1.1 (#34252)
fix(openai): Fixing error that comes up using the Responses API with built-in tools and custom tools (#34136)
docs(core,groq,openai): nits for ref docs (#34243)
test(openai): mark test_structured_output_and_tools flaky (#34223)
chore: update core dep in lockfiles (#34216)
release: (integration packages) 1.1 (#34087)
fix(openai): make GPT-5 temperature validation case-insensitive (#34012)
style(core,langchain-classic,openai): fix griffe warnings (#34074)
chore(docs): fix a few refs syntax errors (#34044)
feat(model-profiles): distribute data across packages (#34024)
chore(openai): skip Azure text completions tests (#34021)
style(langchain): fix some middleware ref syntax (#33988)
release(openai): 1.0.3 (#33981)
refactor(openai): embedding utils and calculations (#33982)
fix(openai): Respect 300k token limit for embeddings API requests (#33668)
fix(langchain, openai): backward compat for response_format (#33945)
fix(langchain, openai): fix create_agent / response_format for Responses API (#33939)
chore: update README.md files (#33919)
test(groq,openai): add retries for flaky tests (#33914)
feat: The response.incomplete event is not handled when using stream_mode=['messages'] (#33871)
style: some cleanup (#33857)
chore: attribute package change versions (#33854)
style: refs pass (#33813)
fix(standard-tests): update multimodal tests (#33781)
release(openai): 1.0.2 (#33769)
test(openai): mark tests flaky (#33750)
fix(openai): replace pytest.warns(None) with warnings.catch_warnings in ChatOpenAI test to resolve TypeError . Resolves issue #33705 (#33741)
fix(openai): update model used in test (#33733)
style: more refs fixes (#33730)
style(core): more cleanup all around (#33711)
fix: links (#33691)
fix(openai): update test (#33659)
feat(openai): add unit test for streaming error in _generate (#33134)
release(openai): 1.0.1 (#33624)
feat(openai): add callable support for openai_api_key parameter (#33532)
chore: update pyproject.toml url entries (#33587)
release(openai): 1.0.0 (#33578)
docs: fix Anthropic, OpenAI docstrings (#33566)
fix: support python 3.14 in various projects (#33575)
docs: more fixes for refs (#33554)
chore: more sweeping (#33533)
feat(openai): Populate OpenAI service tier token details (#32721)
style: more work for refs (#33508)
feat(openai): Add openai moderation middleware (#33492)
docs: update package READMEs (#33488)
fix(docs): Fix several typos and grammar (#33487)
fix(openai): conditionally skip test (#33431)
style: llm -> model (#33423)
style: .. code-block:: admonition translations (#33400)
style: address Sphinx double-backtick snippet syntax (#33389)
chore: update Sphinx links to markdown (#33386)
fix(openai): update file index key name (#33350)
style: remove more Optional syntax (#33371)
chore: drop UP045 (#33362)
style: monorepo pass for refs (#33359)
chore(openai): base.py ref pass (#33355)
chore: enrich pyproject.toml files with links to new references, others (#33343)
chore: clean up pyproject.toml files, use core a7 (#33334)
release(openai): 1.0.0a4 (#33316)
fix(standard-tests,openai): minor fix for Responses API tests (#33315)
fix(core,openai,anthropic): delegate to core implementation on invoke when streaming=True (#33308)
fix(openai): disable stream_usage in chat completions if OPENAI_BASE_URL is set (#33298)
chore(openai): (v1) fix sort order of mcp call keys (#33295)
feat(openai): enable stream_usage when using default base URL and client (#33205)
chore(infra): pdm -> hatchling (#33289)
style: drop target-version = "py39" for OpenAI, Anthropic, HuggingFace (#33287)
fix(anthropic,openai): fix tests (#33257)
chore: delete deprecated items (#33192)
docs: v1 docs updates (#33173)
release: v1.0.0 (#32567)
fix(openai): bump min core version (#33188)
feat: (core, standard-tests) support PDF inputs in ToolMessages (#33183)
release(openai): 0.3.34 (#33169)
feat(openai): support openai sdk 2.0 (#33168)
fix(openai): remove future.annotations import from test files (#33144)
docs: more standardization (#33124)
docs: standardize .. code-block directive usage (#33122)
test(openai): raise token limit for o1 test (#33118)
chore: bump ruff version to 0.13 (#33043)
style: repo linting pass (#33089)
fix: lift openai version constraints across packages (#33088)
fix(openai): pin upper bound to prevent Pydantic 2.7.0 issues (#33038)
test: use .get() for safe ls_params access (#33034)
chore: update pyproject.toml files, remove codespell (#33028)
feat(openai): add max_tokens to AzureChatOpenAI (#32959)
feat(core): allow overriding ls_model_name from kwargs (#32541)
revert: "chore: remove ruff target-version" (#32895)
fix(openai): update AzureOpenAIEmbeddings validation logic for openai_api_base (#31782)
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)
release(openai): 0.3.32 (#32691)
fix(openai): Always add raw response object to OpenAI client errors for invoke (#32655)
docs(docs): fixed typos in documentations (#32661)
release(openai): 0.3.31 (#32646)
fix(openai): allow temperature parameter for gpt-5-chat models (#32624)
fix(openai): construct responses api input (#32557)
fix(openai): Chat Message Annotations defaults to [ ] if not list or None (#32614)
release(openai): 0.3.30 (#32515)
feat(openai): officially support verbosity (#32470)
feat: port various nit changes from wip-v0.4 (#32506)
release(openai): release 0.3.29 (#32463)
chore: formatting across codebase (#32466)
fix(openai): revert some changes (#32462)
feat(openai): minimal and verbosity (#32455)
fix(openai): don't crash get_num_tokens_from_messages on gpt-5 (#32451)
feat(openai): custom tools (#32449)
test(openai): add tests for prompt_cache_key parameter and update docs (#32363)
fix(openai): add in output_text (#32450)
chore(openai): bump openai sdk (#32322)
fix(openai): add max_retries parameter to ChatOpenAI for handling 503 capacity errors (#32286)
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)
docs(openai): add comprehensive documentation and examples for extra_body + others (#32149)
release(openai): 0.3.28 (#32015)
fix(openai): support acknowledged safety checks in computer use (#31984)
chore: [openai] bump sdk (#31958)
docs(xai): update for Grok 4 (#31953)
ruff: more rules across the board & fixes (#31898)
fix: automatically fix issues with ruff (#31897)
langchain-openai[patch]: Add ruff bandit rules to linter (#31788)
openai: release 0.3.27 (#31769)
openai[fix]: Correctly set usage metadata for OpenAI Responses API (#31756)
docs: updates from langchain-openai 0.3.26 (#31764)
openai[patch]: update cassette (#31752)
openai[patch]: update test following release of openai 1.92 (#31751)
openai: release 0.3.26 (#31749)
openai[patch]: allow specification of output format for Responses API (#31686)
openai[patch]: fix bug where function call IDs were not populated (#31735)
openai[patch]: add attribute to always use previous_response_id (#31734)
openai[patch]: fix dropping response headers while streaming / Azure (#31580)
openai: release 0.3.25 (#31702)
openai[patch]: fix client caching when request_timeout is specified via httpx.Timeout (#31698)
openai[patch]: Improve error message when response type is malformed (#31619)
openai[patch]: include 'type' key internally when streaming reasoning blocks (#31661)
openai[patch]: route to Responses API if relevant attributes are set (#31645)
openai: release 0.3.24 (#31642)
openai[patch]: add Responses API attributes to BaseChatOpenAI (#30329)
openai[patch]: support Responses streaming in AzureChatOpenAI (#31641)
openai[patch]: refactor handling of Responses API (#31587)
openai: release 0.3.23 (#31604)
openai[patch]: handle annotation change in openai==1.82.0 (#31597)
[OpenAI]: Encoding Model (#31402)
openai: release 0.3.22 (#31542)
openai[patch]: accommodate change in image generation API (#31522)
openai[patch]: VCR some tests (#31524)
openai[release]: 0.3.21 (#31519)
openai[patch]: pass through with_structured_output kwargs (#31518)
openai[patch]: update with_structured_outputs docstring (#31517)
openai: release 0.3.20 (#31515)
openai[patch]: guard against None text completions in BaseOpenAI (#31514)
openai[patch]: clarify docs on api_version in docstring for AzureChatOpenAI (#31502)
openai: release 0.3.19 (#31466)
openai[minor]: add image generation to responses api (#31424)
infra: remove use of --vcr-record=none (#31452)
standard-tests: migrate to pytest-recording (#31425)
infra: drop azure from streaming benchmarks (#31421)
openai: cache httpx client (#31260)
standard-tests: add benchmarks (#31302)
openai[patch]: add test case for code interpreter (#31383)
openai: release 0.3.18 (#31320)
openai[patch]: relax Azure llm streaming callback test (#31319)
openai[patch]: support built-in code interpreter and remote MCP tools (#31304)
openai[patch]: run tokenize in background thread in async embedding invocations (#31312)
openai[patch]: update tests for strict schemas (#31306)
openai[patch]: reduce tested constraints on strict schema adherence for Responses API (#31290)
openai[patch]: compat with Bedrock Converse (#31280)
openai: release 0.3.17 (#31246)
openai: support runtime kwargs in embeddings (#31195)
openai: ignore function_calls if tool_calls are present (#31198)
partners: (langchain-openai) total_tokens should not add 'Nonetype' t… (#31146)
Replace deprecated .dict() with .model_dump() for Pydantic v2 compatibility (#31107)
openai: release 0.3.16 (#31100)
openai[patch]: format system content blocks for Responses API (#31096)
core, openai[patch]: prefer provider-assigned IDs when aggregating message chunks (#31080)
openai[patch]: propagate service_tier to response metadata (#31089)
openai[patch]: release 0.3.15 (#31087)
packaging: remove Python upper bound for langchain and co libs (#31025)
langchain_openai: clean duplicate code for openai embedding. (#30872)
docs: document OpenAI flex processing (#31023)
openai[patch]: add explicit attribute for service tier (#31005)
openai[patch]: delete redundant tests (#31004)
openai: fix streaming reasoning without summaries (#30999)
core, standard-tests: support PDF and audio input in Chat Completions format (#30979)
core, openai: support standard multi-modal blocks in convert_to_openai_messages (#30968)
openai[patch]: remove xfails from image token counting tests (#30963)
Support of openai reasoning summary streaming (#30909)
partners/openai: OpenAIEmbeddings not respecting chunk_size argument (#30757)
openai: release 0.3.14 (#30908)
standard-tests, openai[patch]: add support standard audio inputs (#30904)
multiple: permit optional fields on multimodal content blocks (#30887)
docs: document OpenAI reasoning summaries (#30882)
openai: release 0.3.13 (#30858)
multiple: multi-modal content blocks (#30746)
openai[patch]: update imports in test (#30828)
partners[lint]: run pyupgrade to get code in line with 3.9 standards (#30781)
Clean up numpy dependencies and speed up 3.13 CI with numpy>=2.1.0 (#30714)
openai[patch]: make computer test more reliable (#30672)
openai[patch]: upgrade tiktoken and fix test (#30621)
openai: release 0.3.12 (#30616)
openai[patch]: support structured output and tools (#30581)
openai[patch]: ignore file blocks when counting tokens (#30601)
langchain-openai: Support token counting for o-series models in ChatOpenAI (#30542)
docs,langchain-community: Fix typos in docs and code (#30541)
openai: release 0.3.11 (#30503)
openai[patch]: support streaming token counts in AzureChatOpenAI (#30494)
infra: handle flaky tests (#30501)
openai[patch]: attempt to make test less flaky (#30463)
openai[patch]: bump openai sdk (#30461)
openai: release 0.3.10 (#30460)
openai[patch]: support multi-turn computer use (#30410)
openai[patch]: trace strict in structured_output_kwargs (#30425)
Fix typo: change 'ben' to 'be' in comment (#30358)
openai[patch]: release 0.3.9 (#30325)
openai[patch]: support additional Responses API features (#30322)
openai[patch]: support structured output via Responses API (#30265)
openai[patch]: support Responses API (#30231)
standard-tests, openai: bump core (#30202)
openai[patch]: release 0.3.8 (#30164)
core[patch]: update structured output tracing (#30123)
core, openai, standard-tests: improve OpenAI compatibility with Anthropic content blocks (#30128)
docs: Fix typo in code samples for max_tokens_for_prompt (#30088)
openai[patch]: add unit test (#30022)
openai[patch]: release 0.3.7 (#29967)
openai[patch]: set global ssl context (#29932)
core, openai[patch]: support serialization of pydantic models in messages (#29940)
openai[patch]: update system role to developer for o-series models (#29785)
openai: Update reasoning_effort arg documentation (#29897)
core: basemessage.text() (#29078)
openai[patch]: release 0.3.6 (#29824)
openai[patch]: enable streaming for o1 (#29823)
multiple: fix uv path deps (#29790)
Fix #29759: Use local chunk_size for looping in embed_documents (#29761)
openai[patch]: release 0.3.5 (#29740)
openai, deepseek: make _convert_chunk_to_generation_chunk an instance method (#29731)
Community/Partner: Adding Azure community and partner user agent to better track usage in Python (#29561)
openai[patch]: make parallel_tool_calls explicit kwarg of bind_tools (#29669)
openai: Removed tool_calls from completion chunk after other chunks have already been sent. (#29649)
openai: release 0.3.4 (#29652)
openai[patch]: add test for message.name (#29651)
infra: add UV_FROZEN to makefiles (#29642)
infra: migrate to uv (#29566)
partners/openai: release 0.3.3 (#29490)
multiple: structured output tracing standard metadata (#29421)
openai[patch]: fix int test (#29395)
openai[patch]: detect old models in with_structured_output (#29392)
Community: fix missing f-string modifier in oai structured output parsing error (#29326)
openai[patch]: fix Azure LLM test (#29302)
openai[patch]: release 0.3.1 (#29301)
anthropic[patch]: support parallel_tool_calls (#29257)
docs: oai api ref nit (#29210)
openai: disable streaming for o1 by default (#29147)
docs: Update openai README.md (#29146)
openai[minor]: release 0.3 (#29100)
openai[patch]: support streaming with json_schema response format (#29044)
Revert "integrations[patch]: remove non-required chat param defaults" (#29048)
integrations[patch]: remove non-required chat param defaults (#26730)
patch openai json mode test (#28831)
xfail openai image token count test (#28828)
openai[patch]: Release 0.2.14 (#28826)
openai[patch]: type reasoning_effort (#28825)
openai[patch]: Release 0.2.13 (#28800)
core[patch], openai[patch]: Handle OpenAI developer msg (#28794)
json mode standard test (#25497)
openai[patch]: release 0.2.12 (#28633)
openai[patch]: bump minimum SDK version (#28632)
multiple: fix xfailed signatures (#28597)
openai[patch]: add standard tests for embeddings (#28540)
openai: dont populate logit_bias if None (#28482)
partners/openai: release 0.2.11 (#28461)
openai: set logit_bias to none instead of empty dict by default (#28460)
openai[patch]: bump core dep (#28361)
openai[patch]: Release 0.2.10 (#28360)
openai[patch]: use max_completion_tokens in place of max_tokens (#26917)
groq,openai,mistralai: fix unit tests (#28279)
openai[patch]: unskip test and relax tolerance in embeddings comparison (#28262)
multiple: push deprecation removals to 1.0 (#28236)
standard-tests: rename langchain_standard_tests to langchain_tests, release 0.3.2 (#28203)
openai: release 0.2.9, o1 streaming (#28197)
multiple: langchain-standard-tests -> langchain-tests (#28139)
openai[patch]: release 0.2.8 (#28062)
anthropic[major]: support python 3.13 (#27916)
openai[patch]: fix azure oai stream check (#28048)
openai[patch]: Release 0.2.7 (#28047)
openai[patch]: default to invoke on o1 stream() (#27983)
openai[patch]: release 0.2.6 (#27924)
integrations[patch]: bump core to 0.3.15 (#27805)
openai[patch]: add check for none values when summing token usage (#27585)
openai[patch]: fix schema formatting util (#27685)
openai[patch]: Release 0.2.4 (#27652)
openai[patch]: accept json schema response format directly (#27623)
partners/openai + community: Async Azure AD token provider support for Azure OpenAI (#27488)
partners/openai: release 0.2.3 (#27457)
openai: audio modality, remove sockets from unit tests (#27436)
multiple: rely on asyncio_mode auto in tests (#27200)
openai[patch]: default temp=1 for o1 (#27206)
standard-tests[patch]: fix oai usage metadata test (#27122)
openai[patch]: Release 0.2.2 (#27119)
core,integrations[minor]: Dont error on fields in model_kwargs (#27110)
openai: fix None token detail (#27091)
openai[patch]: add usage metadata details (#27080)
openai[patch]: exclude http client (#26891)
openai[patch]: update deprecation target in API ref (#26921)
openai[patch]: increase token limit in azure integration tests (#26901)
openai[patch]: Release 0.2.1 (#26858)
openai[patch]: fix azure integration tests (#26791)
openai[patch]: enable Azure structured output, parallel_tool_calls=Fa… (#26599)
openai: embeddings: supported chunk_size when check_embedding_ctx_length is disabled (#23767)
anthropic[patch]: fix tool call and tool res image_url handling (#26587)
core,openai,grow,fw[patch]: deprecate bind_functions, update chat mod… (#26584)
openai[minor]: Release 0.2.0 (#26464)
multiple: pydantic 2 compatibility, v0.3 (#26443)
openai[patch]: Release 0.1.25 (#26439)
openai[patch]: get output_type when using with_structured_output (#26307)
partners/openai: release 0.1.24 (#26417)
community, openai: support nested dicts (#26414)
fmt
standard-tests[patch]: add Ser/Des test
langchain_openai: Make sure the response from the async client in the astream method of ChatOpenAI is properly awaited in case of "include_response_headers=True" (#26031)
openai[patch]: add back azure embeddings api_version alias
openai[patch]: Upgrade @root_validators in preparation for pydantic 2 migration (#25491)
standard-tests[patch]: test init from env vars (#25983)
docs: ChatOpenAI.with_structured_output nits (#25952)
langchain_openai: Cleanup OpenAIEmbeddings validate_environment. (#25855)
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)
openai[patch]: update core dep (#25502)
core, partners: implement standard tracing params for LLMs (#25410)
openai[patch]: Release 0.1.22 (#25496)
openai[patch]: fix json mode for Azure (#25488)
docs: format oai embeddings docstring (#25448)
docs: remove the extra period in docstring (#25414)
docs: udpated api reference (#25172)
openai[patch]: Update extra to match pydantic 2 (#25382)
openai: Update API Reference docs for AzureOpenAI Embeddings (#25312)
openai[patch]: Docs fix RST formatting in OpenAIEmbeddings (#25293)
openai[patch]: Add API Reference docs to OpenAIEmbeddings (#25290)
docs: Standardize OpenAI Docs (#25280)
openai[patch]: Release 0.1.21 (#25269)
partners: fix of issue #24880 (#25229)
infra: temp skip oai embeddings test (#25148)
openai[patch]: Release 0.1.21rc2 (#25146)
openai[patch]: ChatOpenAI.with_structured_output json_schema support (#25123)
openai[patch]: Release 0.1.21rc1 (#25116)
core[patch], openai[patch]: enable strict tool calling (#25111)
patch[Partners] Unified fix of incorrect variable declarations in all check_imports (#25014)
integrations[patch]: release model packages (#24900)
core[patch], integrations[patch]: convert TypedDict to tool schema support (#24641)
openai[patch]: openai proxy added to base embeddings (#24539)
integration releases (#24725)
openai[small]: Add the new model: gpt-4o-mini (#24594)
openai[patch]: move test (#24552)
standard-tests: add override check (#24407)
all: add release notes to pypi (#24519)
core,groq,openai,mistralai,robocorp,fireworks,anthropic[patch]: Update BaseModel subclass and instance checks to handle both v1 and proper namespaces (#24417)
openai: release 0.1.18 (#24369)
openai[patch]: use model_name in AzureOpenAI.ls_model_name (#24366)
openai: release 0.1.17 (#24361)
openai[patch]: fix key collision and _astream (#24345)
openai: raw response headers (#24150)
openai: remove some params from default serialization (#24280)
openai[patch]: update openai params (#23691)
openai[patch]: Release 0.1.16 (#24202)
integrations[patch]: require core >=0.2.17 (#24207)
standard-tests[patch]: add tests for runnables as tools and streaming usage metadata (#24153)
core[minor], integrations...[patch]: Support ToolCall as Tool input and ToolMessage as Tool output (#24038)
openai: release 0.1.15 (#24097)
openai: Create msg chunk in bg thread (#24032)
infra: update mypy 1.10, ruff 0.5 (#23721)
openai[patch]: Release 0.1.14 (#23782)
openai[patch]: expose model request payload (#23287)
partners[azure]: fix having openai_api_base set for other packages (#22068)
openai[patch]: Release 0.1.13 (#23651)
openai, anthropic, ...: with_structured_output to pass in explicit tool choice (#23645)
docs: azure openai docstring (#23643)
openai[patch]: release 0.1.12 (#23641)
standard-tests: add test for structured output (#23631)
openai[patch]: release 0.1.11 (#23596)
openai[patch]: revert addition of "name" to supported properties for tool messages (#23600)
openai[patch]: add comment (#23595)
partners[openai]: Add name field to tool message to match OpenAI spec (#23551)
partners[openai]: only add stream_options to kwargs if requested (#23552)
openai[patch]: bump sdk version (#23592)
openai[patch]: add support for extra_body (#23404)
[docs]: split up tool docs (#22919)
openai[patch]: Release 0.1.10 (#23410)
openai[patch]: fix tool calling token counting (#23408)
openai[patch]: temp fix ignore lint (#23290)
openai: release 0.1.9 (#23263)
partners[minor]: Fix value error message for with_structured_output (#22877)
infra: add more formatter rules to openai (#23189)
openai[patch]: image token counting (#23147)
openai[patch], standard-tests[patch]: don't pass in falsey stop vals (#23153)
standard-tests[patch]: Update chat model standard tests (#22378)
openai[patch]: add stream_usage parameter (#22854)
[Partner]: Add metadata to stream response (#22716)
partners: fix numpy dep (#22858)
openai: add parallel_tool_calls to api ref (#22746)
openai[patch]: correct grammar in exception message in embeddings/base.py (#22629)
openai, azure: update model_name in ChatResult to use name from API response (#22569)
docs: update anthropic chat model (#22483)
openai: update ChatOpenAI api ref (#22324)
openai[patch]: Release 0.1.8 (#22291)
docs: revamp ChatOpenAI (#22253)
openai: don't override stream_options default (#22242)
openai: read stream_options (#21548)
docs: fix some spelling mistakes caught by newest version of code spell (#22090)
anthropic, openai: cut pre-releases (#22083)
core, partners: add token usage attribute to AIMessage (#21944)
infra: rm unused # noqa violations (#22049)
openai[patch], community[patch]: add pricing and max context window for GPT-4o (#21673)
partners: bump core in packages implementing ls_params (#21868)
core, standard tests, partner packages: add test for model params (#21677)
openai[patch]: fix embedding float precision issue (#21736)
openai[patch]: Release 0.1.7, bump tiktoken 0.7.0 (#21723)
multiple: core 0.2 nonbreaking dep, check_diff community->langchain dep (#21646)
openai: fix loads unit test (#21542)
openai[patch]: Release 0.1.6 (#21236)
openai[patch]: support tool_choice="required" (#21216)
openai[patch]: fix azure open lc serialization, release 0.1.5 (#21159)
openai[patch]: add tests for secret_str for keys (#20982)
openai[patch]: fix special token default behavior (#21131)
multiple: Remove unnecessary Ruff suppression comments (#21050)
openai: release 0.1.4 (#20939)
openai: add unit test (#20931)
mistral, openai: support custom tokenizers in chat models (#20901)
openai[patch]: Allow disablling safe_len_embeddings(OpenAIEmbeddings) (#19743)
patch: remove usage of llm, chat model call (#20788)
docs, multiple: de-beta with_structured_output (#20850)
core, openai: support custom token encoders (#20762)
openai: fix allowed block types (#20636)
mistral, openai: allow anthropic-style messages in message histories (#20565)
update scheduled tests (#20526)
release anthropic, fireworks, openai, groq, mistral (#20333)
openai[patch]: use tool_calls in request (#20272)
openai[patch]: Fix langchain-openai unknown parameter error with gpt-4-turbo (#20271)
infra, multiple: rc release versions (#20252)
openai[patch]: pre-release 0.1.3-rc.1 (#20249)
core[minor], ...: add tool calls message (#18947)
openai[patch]: Release 0.1.2 (#20241)
standard-tests: a standard unit and integration test set (#20182)
openai[patch]: wrap stream code in context manager blocks (#18013)
docs: use standard openai params (#20160)
openai[patch]: remove openai chunk size validation (#19878)
openai[patch]: Partially Revert Update openai chat model to new base class interface (#19871)
openai[patch]: fix azure embedding length check (#19870)
openai[patch]: Update openai chat model to new base class interface (#19729)
openai[patch]: add checking codes for calling AI model get error (#17909)
openai[patch]: perform judgment processing on chat model streaming delta (#18983)
openai[patch]: fix ChatOpenAI model's openai proxy (#19559)
openai[patch]: fix AzureChatOpenAI missing parameter problem (#19258)
openai[patch]: increase max batch size for Azure OpenAI Embeddings API (#19532)
openai[patch]: add test coverage to output (#19462)
openai[patch]: tool use integration test (#19460)
openai[patch]: integration test structured output (#19459)
openai[patch]: release 0.1.1 (#19458)
openai[patch]: [URGENT REGRESSION FIX] Don't fail if tool message already doesn't contain name (#19435)
openai[patch]: fix core min version (#19366)
openai[patch]: fix name param (#19365)
openai[patch]: release 0.1.0, message id and name support (#19363)
openai[patch]: pass message name (#17537)
core[patch]: Add LLM output to message response_metadata (#19158)
openai[patch]: fix async http client (#19164)
core[minor]: generation info on msg (#18592)
core[patch]: support labeled json schema as tools (#18935)
partners[openai]: update the docstring of OpenAI, OpenAIEmbeddings and ChatOpenAI classes (#18908)
openai[patch]: Proper example for AzureOpenAI usage in error message (#17798)
infra: ChatOpenAI unit tests for invoke() and ainvoke() (#18792)
openai[patch]: unit test azure init (#18703)
docs: remove duplicate quote in AzureOpenAIEmbeddings doc (#18315)
langchain_openai[patch]: Invoke callback prior to yielding token (#18269)
openai[patch]: Assign message id in ChatOpenAI (#17837)
openai[patch], mistral[patch], fireworks[patch]: releases 0.0.8, 0.0.5, 0.0.2 (#18186)
langchain[patch], core[patch], openai[patch], fireworks[minor]: ChatFireworks.with_structured_output (#18078)
openai[patch]: refactor with_structured_output (#18052)
openai[patch]: remove numpy dep (#18034)
openai[patch]: Release 0.0.7 (#17993)
core[minor], openai[minor], langchain[patch]: BaseLanguageModel.with_structured_output #17302)
langchain_openai[patch]: fix typos in langchain_openai (#17923)
partners/openai: fix depracation errors of pydantic's .dict() function (reopen #16629) (#17404)
partners: version constraints (#17492)
openai[patch]: relax tiktoken constraint, release 0.0.6 (#17472)
openai[patch]: code cleaning (#17355)
infra: add print rule to ruff (#16221)
infra, multiple: Upgrade configuration for Ruff v0.2.0 (#16905)
langchain[patch]: expose cohere rerank score, add parent doc param (#16887)
openai[patch]: rm tiktoken model warning (#16964)
langchain_openai[patch]: Invoke callback prior to yielding token (#16909)
openai[minor]: change to secretstr (#16803)
openai[patch]: Added annotations support to azure openai (#13704)
openai[patch]: Release 0.0.5 (#16598)
openai[patch]: Explicitly support embedding dimensions (#16596)
openai[patch]: Release 0.0.4 (#16590)
docs: add rag citations page (#16549)
openai[patch]: accept function_call dict in bind_functions (#16483)
core[patch], community[patch], openai[patch]: consolidate openai tool… (#16485)
openai[patch]: Release 0.0.3 (#16289)
core[patch], openai[patch]: Chat openai stream logprobs (#16218)
openai[patch]: clarify azure error (#16157)
partners: add license field (#16117)
openai[patch]: 0.0.2.post1, urls (#15961)
openai[patch]: unit test load (#15624)
openai[patch], docs: readme (#15773)
openai[patch]: v0.0.2 (#15618)
openai: core version, rc1 (#15614)
google-vertexai, openai: release candidate version (#15611)
openai[minor]: implement langchain-openai package (#15503)

Changes since langchain-openrouter==0.2.7

release(openrouter): 0.2.8 (#39658)

chore(model-profiles): refresh model profile data (#39646)

chore(model-profiles): refresh model profile data (#39625)

fix(openrouter): preserve cost metadata in usage chunks (#39338)

chore(model-profiles): refresh model profile data (#39609)

chore(model-profiles): refresh model profile data (#39579)

chore(model-profiles): refresh model profile data (#39430)

chore(model-profiles): refresh model profile data (#39344)

chore(model-profiles): refresh model profile data (#39318)

feat(openrouter): preserve provider in response metadata (#39301)

chore(model-profiles): refresh model profile data (#39299)

chore(model-profiles): refresh model profile data (#39239)

chore(model-profiles): refresh model profile data (#39166)

chore(model-profiles): refresh model profile data (#39141)

chore(model-profiles): refresh model profile data (#39105)

chore(model-profiles): refresh model profile data (#39084)

chore(model-profiles): refresh model profile data (#39050)

chore(model-profiles): refresh model profile data (#39036)

Changes since langchain-core==1.5.4

release(core): 1.5.5 (#39655)

fix(core): make abatch_iterate consistent with batch_iterate for None and zero size (#39367)

fix(core): respect pydantic aliases when validating tool inputs (#39572)

fix(core): issues in merging chunks (#39535)

fix(core): handle v1 base model validation in async path (#39576)

fix(core): handle tool descriptions for infer_schema=False (#39573)

fix(core): clear usage metadata callback on exceptions in context manager (#39616)

fix(core): handle falsy LLM and chat model caches (#39283)

chore(core): add httpx as an explicit dep (#39612)

fix(core): preserve non-str/non-dict items in DictPromptTemplate list values (#39588)

fix(core): raise ValueError when explicit tool_outputs length mismatches tool_calls in tool_example_to_messages (#39142)

fix(core): guard malformed Anthropic content blocks (#38670)