本页还没有译文
本页 92 段还没译成한국어,下面显示的是英文原文。Release policy
还没翻译译自英文(docs.langchain.com)本页 92 段都还没译
This page explains the LangChain and LangGraph release policies. Click on the tabs below to view the release policies for each:
LangChainLangGraphDeep Agents
The LangChain ecosystem is composed of different component packages (e.g., langchain-core, langchain, partner packages, etc.)
Release cadence
With the release of LangChain 1.0, minor releases (e.g., from 1.0.x to 1.1.0) of langchain and langchain-core follow semantic versioning and may be released frequently. Minor releases contain new features and improvements but do not include breaking changes.
Patch versions are released frequently, up to a few times per week, as they contain bug fixes and minor improvements.
API stability
The development of LLM applications is a rapidly evolving field, and we are constantly learning from our users and the community. As such, we expect that the APIs in langchain and langchain-core will continue to evolve to better serve the needs of our users.
With LangChain 1.0's adoption of semantic versioning:
- Breaking changes to the public API will only occur in major version releases (e.g.,
2.0.0) - Minor version bumps (e.g.,
1.0.0to1.1.0) add new features without breaking changes - Patch version bumps (e.g.,
1.0.0to1.0.1) contain bug fixes and minor improvements
We will generally try to avoid making unnecessary changes, and will provide a deprecation policy for features that are being removed.
Stability of other packages
The stability of other packages in the LangChain ecosystem may vary:
- Partner packages maintained by LangChain (such as
langchain-openaiandlangchain-anthropic) follow semantic versioning and are expected to be stable post 1.0. Other partner packages may follow different stability and versioning policies, and users should refer to the documentation of those packages for more information.
Deprecation policy
We will generally avoid deprecating features until a better alternative is available.
With LangChain 1.0's semantic versioning approach, deprecated features will continue to work throughout the entire 1.x release series. Breaking changes, including the removal of deprecated features, will only occur in major version releases (e.g., 2.0).
When a feature is deprecated in langchain or langchain-core, we will:
- Clearly mark it as deprecated in the code and documentation
- Provide migration guidance to the recommended alternative
- Provide security updates for the deprecated feature through all 1.x minor releases
In some situations, we may allow deprecated features to remain in the code base even longer if they are not causing maintenance issues, to further reduce the burden on users.
Long-term support (LTS)
LangChain follows a long-term support (LTS) policy to provide stability for production applications:
Release status definitions
Packages are marked with one of the following statuses:
- ACTIVE: Current active development, includes bug fixes, security patches, and new features
- MAINTENANCE: Receives all security patches and critical bug fixes, but no new features
Current LTS releases
LangChain 1.0 is designated as an LTS release:
- Status: ACTIVE until the release of 2.0
- Support period: After 2.0 is released, 1.0 will enter MAINTENANCE mode for at least 1 year
- Semver compliance: Users can upgrade between minor versions (e.g., 1.0 to 1.1) without breaking changes
Legacy version support
LangChain 0.3:
- Status: MAINTENANCE mode
- Support period: Until December 2026
- Support includes: Security patches and critical bug fixes
LangGraph follows a structured release policy to ensure stability and predictability for users building production applications.
Release cadence
We expect to space out major releases by at least 6-12 months to provide stability for production applications.
Minor releases are typically released every 1-2 months with new features and improvements.
Patch releases are released as needed, often weekly, to address bugs and security issues.
API stability
Stable APIs
All APIs without special prefixes are considered stable and ready for production use. We maintain backward compatibility for stable features within a major version.
Beta features
Features marked as beta in the documentation are:
- Feature-complete and tested
- Safe for production use with the understanding they may change
- Subject to minor API adjustments based on user feedback
Experimental features
Features marked as experimental or alpha:
- Are under active development
- May change significantly or be removed
- Should be used with caution in production
Internal APIs
APIs prefixed with underscore (_) or explicitly marked as internal:
- Are not part of the public API
- May change without notice
- Should not be used directly
Deprecation policy
When deprecating features:
- Deprecation Notice: Features are marked as deprecated with clear migration guidance
- Grace Period: Deprecated features remain functional for at least one minor version
- Removal: Features are removed only in major version releases
- Migration Support: We provide migration guides and, when possible, automated tools
Platform compatibility
Python support
- We support Python versions that are actively maintained by the Python Software Foundation
- Python version requirements may change only in major releases
- Currently requires Python 3.10 or later
Breaking changes
Breaking changes are only introduced in major versions and include:
- Removal of deprecated APIs
- Changes to required parameters
- Changes to default behavior that affect existing applications
- Minimum Python/Node.js version updates
Migration support
For major version upgrades, we provide:
- Comprehensive migration guides
- Automated migration scripts when feasible
- Extended support period for the previous major version
- Clear documentation of all breaking changes
Long-term support (LTS)
LangGraph follows a long-term support (LTS) policy to provide stability for production applications:
Release status definitions
Packages are marked with one of the following statuses:
- ACTIVE: Current active development, includes bug fixes, security patches, and new features
- MAINTENANCE: Receives all security patches and critical bug fixes, but no new features
Current LTS releases
LangGraph 1.0 is designated as an LTS release:
- Status: ACTIVE until the release of 2.0
- Support period: After 2.0 is released, 1.0 will enter MAINTENANCE mode for at least 1 year
- Semver compliance: Users can upgrade between minor versions (e.g., 1.0 to 1.1) without breaking changes
Legacy version support
LangGraph 0.4:
- Status: MAINTENANCE mode
- Support period: Until December 2026
- Support includes: All security patches and critical bug fixes
See also
- Versioning - Version numbering and support details
- Releases - Version-specific release notes and migration guides
Deep Agents (deepagents) is a rapidly evolving package for building advanced agent architectures.
Pre-1.0 status
Deep Agents is currently in pre-1.0 development. While we minimize breaking changes when possible, the API may evolve as we incorporate learnings from the community and adapt to the rapidly changing landscape of agentic applications.
Release cadence
Minor releases (e.g., 0.1.0 to 0.2.0) may contain new features and potentially breaking changes as the package matures.
Patch releases (e.g., 0.1.0 to 0.1.1) contain bug fixes and minor improvements without breaking changes.
API stability
As a pre-1.0 package:
- APIs may change between minor versions
- We aim to minimize disruption by providing clear migration guidance when changes occur
- Features marked as
experimentaloralphaare subject to more significant changes
Deprecation policy
When deprecating features in Deep Agents:
- We provide deprecation warnings in advance when possible
- Migration guidance is included in release notes
- Deprecated features are removed in subsequent minor releases
Path to 1.0
Deep Agents will reach 1.0 when:
- Core APIs have stabilized based on community feedback
- The package has been battle-tested in production environments
- Breaking changes are no longer expected for core functionality
After 1.0, Deep Agents will follow the same semantic versioning and LTS policies as LangChain and LangGraph.
See also
- Versioning - Version numbering and support details