LangChain
AI & LLMs🦜🔗 The platform for reliable agents.
Release History
langchain==1.3.13langchain-openai==1.3.5langchain-fireworks==1.4.4langchain-openai==1.3.42 fixesVersion 1.3.4 of langchain-openai primarily focuses on bug fixes, including suppressing Pydantic warnings and improving async loop handling in core components.
langchain==1.3.123 fixesLangChain version 1.3.12 primarily focuses on bug fixes, including improved interrupt propagation in retries and better handling of Anthropic cache markers during fallbacks.
langchain-core==1.4.96 fixesThis release focuses on stability and bug fixes within the core library, addressing issues in output parsing, error handling, and async context usage, alongside dependency bumps.
langchain-mistralai==1.1.63 featuresThis release for langchain-mistralai surfaces citation metadata and adds support for stop sequences. Core components received updates for package version tracking in tracing metadata.
langchain-openrouter==0.2.61 fixThis release of langchain-openrouter primarily fixes an issue by adding support for custom HTTP headers via `default_headers`.
langchain-openrouter==0.2.52 fixesThis release of langchain-openrouter focuses on minor bug fixes, specifically deduplicating finish metadata and stripping reasoning IDs from responses.
langchain-anthropic==1.4.82 fixesThis release primarily contains bug fixes for the Anthropic integration and core error handling, alongside a minor dependency bump for langgraph-checkpoint.
langchain-fireworks==1.4.31 fix1 featureThis release primarily focuses on dependency updates across various components, including langsmith, vcrpy, and pytest, alongside introducing package version tracking for tracing metadata.
langchain-openrouter==0.2.42 fixes2 featuresThis release (0.2.4) for langchain-openrouter bumps its underlying `openrouter` dependency and introduces feature parity for parallel tool calls. It also includes numerous dependency updates and infrastructure chores across the repository.
langchain==1.3.111 fixThis release fixes an issue in `ProviderStrategy` regarding tool strictness for OpenAI models and updates several dependencies.
langchain-anthropic==1.4.71 fixThis release primarily contains dependency bumps for vcrpy and langsmith, along with a hotfix to regenerate the anthropic cassette and minor documentation clarifications.
langchain-openai==1.3.33 fixesThis patch release for langchain-openai primarily focuses on bug fixes related to response handling, tool strictness, and dependency updates for langsmith and vcrpy.
langchain==1.3.102 fixesThis release includes minor dependency bumps, updates to summary formatting, and fixes for provider snapshot detection across several components.
langchain-core==1.4.8Breaking2 fixesThis release bumps several dependencies, removes support for Python < 3.10 in langchain-classic, and includes performance improvements and bug fixes related to streaming events.
langchain-openai==1.3.2This release (1.3.2) for langchain-openai contains minor updates since version 1.3.1.
langchain-openai==1.3.13 fixes2 featuresThis release introduces package version tracking in tracing metadata and fixes issues related to streamed tool calls and structured output fallbacks. It also updates internal testing configurations.
langchain-core==1.4.72 fixesVersion 1.4.7 of langchain-core primarily addresses bug fixes, including restoring Pydantic v1 compatibility in tools/runnable, and updates minor dependencies.
langchain-anthropic==1.4.62 fixes1 featureThis release focuses on infrastructure improvements, including a mypy bump and unified type-checking, alongside a fix to confine file-search results in Anthropic integrations and new version tracking for tracing metadata.
langchain==1.3.92 fixesThis release includes minor fixes for LangChain and the Anthropic integration, specifically addressing file-search confinement and tightening allowed prefixes.
langchain==1.3.82 fixes1 featureThis release primarily addresses minor bug fixes related to async middleware and structured output fallbacks, alongside adding type overloads for `create_agent`.
langchain-core==1.4.61 fix1 featureVersion 1.4.6 introduces package version tracking for tracing metadata and fixes an issue with normalizing streamed tool calls from OpenAI.
langchain-model-profiles==0.0.67 fixes7 featuresThis release focuses heavily on dependency updates, particularly within `langchain-model-profiles`, and introduces content-block-centric streaming (v2) in the core library. Bug fixes address issues in Anthropic, Fireworks, and schema drift warnings in model profiles.
langchain-core==1.4.52 fixes1 featureLangchain core version 1.4.5 introduces validation for streaming tool call chunks in tests and fixes an issue with async tracer fallback in sync contexts.
langchain-mistralai==1.1.51 fixThis release primarily focuses on dependency updates and minor fixes, including supporting content block tokens in callbacks and bumping several internal dependencies.
langchain-anthropic==1.4.51 fixThis release primarily focuses on bug fixes, specifically supporting content block tokens in callbacks, and includes minor dependency updates.
langchain-classic==1.0.81 fixVersion 1.0.8 of langchain-classic includes a fix for content block tokens in callbacks and updates several internal dependencies to newer versions.
langchain-core==1.4.43 fixesThis hotfix release bumps lockfiles and includes several bug fixes related to callbacks, tool error handling, and typing improvements in langchain-core.
langchain==1.3.71 featureVersion 1.3.7 introduces the `ProviderToolSearchMiddleware` and prepares for future compatibility by marking legacy views for removal.
langchain-groq==1.1.33 fixes4 featuresThis release focuses on dependency updates, infrastructure hardening, and feature enhancements, notably introducing content-block-centric streaming (v2) and Strict Mode for Groq.
langchain==1.3.61 fixThis release (1.3.6) primarily focuses on fixing an issue related to preserving summarization trigger compatibility.
langchain==1.3.51 fix2 featuresThis release introduces AND-capable trigger conditions for `SummarizationMiddleware` and adds support for the `apply_patch` tool in OpenAI. It also updates several underlying dependencies.
langchain-perplexity==1.4.01 fix2 featuresVersion 1.4.0 for langchain-perplexity introduces support for tool binding and improved tool round-trip handling with the Responses-API, alongside a minor dependency update.
langchain-openai==1.3.01 fix1 featureThis release introduces support for the `apply_patch` built-in tool and includes a hotfix to update the minimum required core dependency version for langchain-openai.
langchain-core==1.4.31 featureThis release introduces support for the `apply_patch` built-in tool within OpenAI integrations.
langchain-core==1.4.2This release primarily focuses on deprecating the problematic `dict()` method within the core library.
langchain-perplexity==1.3.22 fixesVersion 1.3.2 for langchain-perplexity primarily addresses serialization issues related to ToolMessage and AIMessage.tool_calls.
langchain-core==1.4.14 fixesThis release focuses on minor fixes, dependency updates across several components, and documentation improvements, including removing Bedrock prevalidation and fixing stream assembly issues.
langchain-deepseek==1.1.07 fixes3 featuresThis release focuses heavily on dependency updates, infrastructure hardening, and enhancements to model profiling, including new fields and stable sorting. It also introduces content-block-centric streaming (v2) to the core.
langchain==1.3.41 fixThis release primarily focuses on improving the guidance provided for Human-In-The-Loop rejection scenarios.
langchain==1.3.32 featuresThis release bumps the dependency on langgraph and introduces new features for subagent execution and middleware configuration in LangChain.
langchain-anthropic==1.4.41 fixThis patch release for langchain-anthropic primarily fixes an issue by normalizing cross-provider tool-call IDs and includes several dependency bumps.
langchain-perplexity==1.3.1This release primarily updates the underlying perplexityai dependency to version 0.34.1.
langchain-fireworks==1.4.21 fixThis release primarily addresses a serialization bug within the fireworks integration by ensuring non-wire keys are stripped from content parts. It also updates internal dependencies.
langchain-perplexity==1.3.01 featureVersion 1.3.0 introduces the `use_responses_api` flag for ChatPerplexity and updates several internal dependencies.
langchain==1.3.22 fixes2 featuresThis release introduces PII redaction and stream transformer registration features, alongside fixes for TodoListMiddleware and glob_search sorting. It also updates several internal dependencies.
langchain-openai==1.2.23 fixesThis release focuses on maintenance, dependency bumps, and minor fixes within the OpenAI integration, including guarding httpx finalizers and correctly sourcing LLM context sizes.
langchain-tests==1.1.91 featureThis release bumps the idna dependency and introduces an enhancement to streaming assertions within the standard-tests library.
langchain-fireworks==1.4.12 fixesVersion 1.4.1 for langchain-fireworks introduces improved resilience by adding retries for API connection errors and setting a default retry limit.
langchain-fireworks==1.4.0Breaking1 fix1 featureVersion 1.4.0 for langchain-fireworks migrates the integration to the fireworks-ai 1.x SDK and introduces proper context overflow error handling. Several minor dependency bumps were also performed.
langchain-tests==1.1.81 fixThis release primarily consists of dependency bumps and maintenance updates, including refreshing stale OpenAI model references and fixing a test assertion regarding model overrides.
langchain==1.3.11 fixThis patch release (1.3.1) addresses a bug related to Bedrock provider aliasing within the summarization token check.
langchain==1.3.01 featureThis release introduces support for version="v3" in event streaming methods for langchain agents.
langchain-core==1.4.034 fixes14 featuresThis release focuses heavily on infrastructure updates, dependency bumps, and significant hardening across security (SSRF) and tracing mechanisms. New features include content-block-centric streaming (v2) and enhanced metadata tracing for LLM invocations.
langchain==1.2.181 fixThis release reverts a previous change regarding agent tagging and introduces deprecations for 'hub' and load/dump limits within langchain-classic.
langchain-core==0.3.861 fixThis release backports a critical path-traversal security fix to the langchain-core v0.3 branch.
langchain==0.3.301 fixThis release backports security hardening for loads/dumps to version 0.3 and deprecates the 'hub' functionality.
langchain-classic==1.0.7This release of langchain-classic deprecates the 'hub' functionality and limits loads/dumps operations.
langchain==1.3.0a224 fixes13 featuresThis release introduces significant updates to streaming protocols (v2 and v3), enhances agent capabilities with new middleware features, and includes numerous dependency bumps and bug fixes across various components.
langchain-classic==1.0.62 fixesThis release primarily focuses on security hardening for manifest loading and restricts deserialization in the classic storage module, while also updating a dependency.
langchain==0.3.292 fixesThis release focuses on security enhancements by restricting deserialization and hardening the `load()` function against untrusted manifests.
langchain-core==0.3.851 fixThis release focuses on security by hardening the `load()` function against untrusted manifests.
langchain-core==1.3.35 fixes1 featureThis release (1.3.3) focuses on hardening security for manifest loading and fixing several internal logic issues, including batch validation and tracer data preservation.
langchain-fireworks==1.3.12 fixesVersion 1.3.1 of langchain-fireworks includes fixes related to API key requirements in embeddings and cleaning up ToolMessage content blocks.
langchain-mistralai==1.1.41 fixThis release primarily addresses a bug fix in the mistralai integration by cleaning up keys in ToolMessage.
langchain-anthropic==1.4.31 fixThis release primarily focuses on refactoring deprecation targets within langchain-classic and includes a minor bug fix for httpx finalizers in the anthropic integration.
langchain-classic==1.0.51 fix1 featureThis release focuses on dependency updates, minor bug fixes in Fireworks, and introduces content-block-centric streaming (v2) in the core library. Deprecation messages have also been refined.
langchain-openrouter==0.2.31 fixThis release (0.2.3) for langchain-openrouter primarily addresses a streaming bug related to reasoning details.
langchain-mistralai==1.1.32 fixes2 featuresThis release for langchain-mistralai adds image input support for human messages and includes several dependency updates and infrastructure improvements across the ecosystem.
langchain-fireworks==1.3.01 featureThe Fireworks integration has been updated to version 1.3.0, introducing the ability to specify a `service_tier` when initializing `ChatFireworks`.
langchain==1.3.0a119 fixes14 featuresThis release includes numerous performance improvements, dependency bumps, and feature additions like content-block-centric streaming (v2) and integration of stream_events(version='v3'). Several bug fixes address token counting, model recognition, and middleware configuration.
langchain-openrouter==0.2.21 fix2 featuresThis release introduces new fields for session tracking in OpenRouter and implements content-block-centric streaming (v2). It also includes several dependency updates and a fix for usage metadata extraction across multiple providers.
langchain-core==1.4.0a235 fixes14 featuresThis release introduces content-block-centric streaming (v2) and support for the v3 stream protocol, alongside numerous bug fixes related to tracing, validation, and dependency updates. Path saving/loading functions for prompts have been deprecated.
langchain-core==1.4.0a134 fixes16 featuresThis release introduces significant feature updates, including content-block-centric streaming (v2) and support for the v3 event protocol. Numerous bug fixes address issues related to tracing, serialization, SSRF hardening, and model wrapper compatibility.
langchain-fireworks==1.2.11 fix1 featureVersion 1.2.1 primarily addresses minor dependency updates and includes a fix for multimodal content block translation in Fireworks, alongside a new content-block-centric streaming feature in core.
langchain==1.2.171 featureThis release introduces a new decision type to the HITL middleware for enhanced control flow.
langchain-perplexity==1.2.02 fixes5 featuresThis release introduces `PerplexityEmbeddings`, overhauls the Perplexity integration with its official SDK, and adds new fields to model profiles. It also includes several dependency bumps, notably for security and stability.
langchain==1.2.161 fix2 featuresThis release introduces content-block-centric streaming (v2) in core and adds agent type tagging for `create_agent` calls. It also includes several dependency updates and performance improvements.
langchain-tests==1.1.71 featureThis release (1.1.7) introduces content-block-centric streaming (v2) and updates several dependencies, including bumping the minimum required versions for core components.
langchain-anthropic==1.4.21 fix1 featureThis release for langchain-anthropic (1.4.2) restores cache control functionality and introduces content-block-centric streaming (v2) in the core library, alongside bumping minimum dependency versions.
langchain-openai==1.2.11 fix1 featureThis hotfix release bumps minimum core dependency versions and introduces content-block-centric streaming (v2) while fixing an issue with the Responses API check.
langchain-core==1.3.21 featureThis release introduces content-block-centric streaming (v2) to the core library.
langchain-fireworks==1.2.05 fixes3 featuresThis release updates dependency versions across the board, refreshes model profile data, and introduces streaming usage metadata support for the Fireworks integration. It also fixes several bugs related to retries and tool binding in Fireworks.
langchain-core==1.3.12 featuresThis release updates dependencies and introduces minor feature enhancements in the core library, including changes to output formatting and tracer metadata handling.
langchain-openai==1.2.01 featureVersion 1.2.0 introduces a fix to prevent silent streaming hangs within ChatOpenAI and includes minor CI and partner integration chore updates.
langchain-openai==1.1.161 fixThis release primarily addresses a bug related to prompt cache retention during streaming operations in langchain-openai.
langchain-openai==1.1.152 fixesThis release of langchain-openai primarily focuses on bug fixes related to streaming responses and Azure profile inference.
langchain-core==1.3.03 fixes1 featureLangchain-core version 1.3.0 introduces new traceable metadata fields for chat models and LLMs, alongside several bug fixes related to SSRF policies and garbage collection support.
langchain-anthropic==1.4.13 fixes2 featuresThis release (1.4.1) introduces support for new Anthropic model features like Opus 4.7 and adaptive thinking, alongside dependency updates and minor bug fixes.
langchain-huggingface==1.2.23 fixes1 featureThis release focuses on hardening HuggingFace integrations by improving repository ID validation and optimizing API calls, alongside updating model profile data and dependencies.
langchain-core==1.3.0a337 fixes12 featuresThis release introduces new features like enhanced metadata tracing, support for ChatBaseten, and multimodal token counting. It also includes numerous bug fixes related to SSRF hardening, model serialization, and tool handling, alongside the deprecation of prompt saving/loading functions due to added path validation.
langchain-classic==1.0.43 fixesThis release of langchain-classic (1.0.4) introduces deprecations, dependency bumps including pytest and cryptography, and fixes related to mypy errors and Azure AI Foundry model providers.
langchain-openai==1.1.141 fixVersion 1.1.14 of langchain-openai fixes an issue related to image token counting by implementing SSRF-safe transport. This release also updates several dependencies.
langchain-text-splitters==1.1.22 fixes1 featureThis release updates numerous dependencies, enforces SSRF-safe transport for URL splitting, and fixes a data loss issue in RecursiveJsonSplitter.
langchain-core==1.2.31This release ports change 36816 to the v1.2 branch, resulting in version 1.2.31 of langchain-core.
langchain-core==1.2.301 fixThis release primarily focuses on hardening private SSRF utilities within the core library.
langchain-openai==1.1.134 fixes1 featureThis release primarily focuses on bug fixes related to API response handling, token extraction, and User-Agent overrides, alongside adding a feature to impute placeholder filenames for OpenAI file inputs.
langchain-core==1.2.291 fixThis release updates langchain-core to version 1.2.29 and incorporates a fix from PR #36725.
Common Errors
ModuleNotFoundError4 reportsThis error usually means the Langchain library is not installed, or the installed version is outdated and missing the module. Fix it by installing or upgrading Langchain using pip: `pip install langchain --upgrade`. If you are using poetry: `poetry add langchain`
ImportError3 reportsImportError in Langchain often arises when attempting to use a module or tool that hasn't been explicitly registered within Langchain's `load_tools` function or when a necessary dependency is missing. To fix this, ensure that the tool or module is properly registered with a string identifier in `load_tools` and that all required dependencies for the tool are installed in your environment (e.g., using `pip install`). Verify the correct spelling and capitalization of the tool's name during registration and when loading it.
UnicodeDecodeError3 reportsUnicodeDecodeError usually arises when trying to decode a file or string with an encoding different from the one it was encoded with (often when the default system encoding is not UTF-8). To fix this, explicitly specify `encoding='utf-8'` when opening, reading, writing, saving, or decoding files and strings, especially in methods like `save()` that handle serialization. This ensures consistent handling of Unicode characters regardless of the system's default encoding.
NotImplementedError3 reportsThe "NotImplementedError" in LangChain often arises when a class inherits from an abstract base class (like `StructuredTool`) and doesn't implement a required method (like `_run`). To fix this, either implement the missing synchronous `_run` method or, if only an asynchronous `_arun` method exists, provide a synchronous implementation that calls `asyncio.run(_arun(...))` or use `sync_to_async` if direct use of `asyncio.run` is not suitable for the execution environment.
BadRequestResponseError2 reportsBadRequestResponseError in Langchain often arises from exceeding the allowed token limit or malformed requests sent to the LLM provider during multi-turn conversations, especially when streaming. To fix this, either reduce the prompt size by summarizing or truncating earlier conversation turns, or ensure proper formatting and encoding of the request payload, especially when handling streamed responses containing structured data.
OutputParserException2 reportsOutputParserException in Langchain often arises from mismatches between the LLM's generated output and the expected format defined by your OutputParser. To fix it, carefully inspect the LLM output for inconsistencies (incorrect field names, wrong data types, missing fields), and adjust either your prompt to guide the LLM to produce the correct format or modify your OutputParser to be more tolerant of variations. Consider adding error handling, such as a try-except block around the parser, to gracefully manage parsing failures.
Related AI & LLMs Packages
AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
Get up and running with OpenAI gpt-oss, DeepSeek-R1, Gemma 3 and other models.
The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.
LLM inference in C/C++
GPT4All: Run Local LLMs on Any Device. Open-source and available for commercial use.
A high-throughput and memory-efficient inference and serving engine for LLMs
Subscribe to Updates
Get notified when new versions are released