Skip to content

Releases: LLMPages/onesdk

Release v0.0.7 - Bug Fix for Initialization

03 Jan 00:59
d409f8c
Compare
Choose a tag to compare

🐛 Bug Fixes

  • Fixed incorrect initialization in __init__.py
    • Corrected the module import structure
    • Ensured proper initialization of package components
    • Fixed potential circular import issues

💡 Improvements

  • Improved code organization in __init__.py
  • Enhanced package initialization clarity
  • Better module exposure at package level

🛠️ Technical Details

Modified __init__.py to properly expose package components and ensure correct initialization sequence. This fixes issues where some modules were not properly accessible or initialized in the correct order.

📝 Additional Notes

  • No breaking changes
  • Backwards compatible with previous versions
  • Recommended update for all users

Please verify and upgrade to ensure proper package functionality.

How to upgrade:

pip install llm-onsdk --upgrade

Version 0.0.6 - Comprehensive API Documentation Enhancement

27 Dec 16:56
8a09b5b
Compare
Choose a tag to compare

This minor release focuses on improving the documentation across all supported API integrations, enhancing readability and maintainability without changing core functionality.

What's New

  • Comprehensive docstrings added to all methods in each API class (including OpenAI, Anthropic, Cohere, Wenxin, Qwen, MiniMax, Ollama, and others)
  • Improved explanations for key implementation details across all providers
  • Enhanced inline comments for complex logic in all modules

Improvements

  • Better developer experience when working with various LLM APIs
  • Increased code maintainability for future updates
  • Improved alignment with best practices for code documentation
  • Consistent documentation style across all API integrations

Developer Notes

  • No functional changes or logic modifications in this release
  • All existing integrations and usage patterns remain unchanged
  • Developers are encouraged to refer to the updated docstrings for more detailed information about each method in all API classes

Compatibility

  • This update is fully backwards compatible with version 0.0.5
  • No changes to API signatures or behavior for any provider

Affected Modules

  • OpenAI API
  • Anthropic API
  • Cohere API
  • Wenxin (Baidu) API
  • Qwen API
  • MiniMax API
  • Ollama API
  • Base API and utility modules

Future Plans

  • Upcoming releases may include performance optimizations and new features building upon this improved documentation
  • Potential addition of new API integrations in future versions

We recommend all users to upgrade to version 0.0.6 to benefit from the improved documentation across all supported APIs, which will aid in better understanding and utilization of the various LLM integrations.

As always, we appreciate your feedback and contributions. If you encounter any issues or have suggestions for any of the API integrations, please open an issue on our GitHub repository.

Thank you for using our unified LLM API integration library!

v0.0.5 - Directory Structure Refactor

27 Dec 16:31
8822b98
Compare
Choose a tag to compare

Description:

We are pleased to announce the release of llm_onesdk version 0.0.5. This release focuses on improving the package structure and resolving import issues.

Key Changes:

  • Reorganized the project directory structure for better organization and maintainability
  • Updated import statements in test files to resolve relative import issues
  • Modified setup.py to correctly reflect the new package structure

Improvements:

  • Enhanced package organization for easier navigation and development
  • Resolved import issues, improving the overall stability of the package
  • Updated packaging configuration for more reliable distribution

To upgrade to this version, please run:

pip install --upgrade llm_onesdk

We recommend all users to upgrade to this version to benefit from the improved structure and resolved import issues.

As always, we appreciate your feedback and contributions. If you encounter any issues or have suggestions, please open an issue on our GitHub repository.

Thank you for using llm_onesdk!

OneSDK v0.0.4: Comprehensive Enhancement and Multi-Provider Optimization

26 Dec 16:34
3834776
Compare
Choose a tag to compare

We are excited to announce the release of OneSDK v0.0.4, featuring significant improvements across the board. This update brings enhanced functionality, expanded provider support, and improved documentation.

New Features and Improvements:

  1. API Enhancements:

    • Optimized API implementations for multiple providers, including Anthropic, Qwen, Doubao, Kimi, MiniMax, OpenAI, and Wenxin.
    • Standardized error handling across all APIs for improved reliability and consistency.
    • Implemented additional provider-specific features such as multimodal support, knowledge base management, and context caching.
  2. Extended Provider Support:

    • Added preliminary support for Ollama, Gemini, and Cohere (documentation provided, full testing pending).
    • Expanded capabilities for existing providers, including new methods and improved integration.
  3. Documentation Overhaul:

    • Updated main README with a comprehensive provider support table, offering clear insight into each provider's capabilities.
    • Enhanced individual provider documentation in the docs/ directory for improved clarity and completeness.
    • Revised core SDK method descriptions for consistency across all providers.
  4. Testing Improvements:

    • Added corresponding test files for each optimized API, significantly increasing test coverage and reliability.
  5. Package Management:

    • Optimized setup.py for better dependency handling and package configuration.
  6. Logging and Debugging:

    • Improved logging capabilities for easier troubleshooting and debugging.
  7. Performance Optimizations:

    • Enhanced overall performance and reduced latency in API calls.

Known Issues:

  • The APIs for Ollama, Gemini, and Cohere are still in development and may not be fully functional or optimized.

Upgrading:
To upgrade to OneSDK v0.0.4, use pip:

pip install --upgrade llm_onesdk

We recommend reviewing the updated documentation for any changes that might affect your existing implementations.

Future Plans:

  • Complete testing and optimization for Ollama, Gemini, and Cohere APIs.
  • Further enhance multi-modal capabilities across supported providers.
  • Implement additional features based on user feedback and provider updates.

We appreciate your continued support and feedback. If you encounter any issues or have suggestions, please open an issue on our GitHub repository.

Thank you for using OneSDK!

llm_onesdk v0.0.3 Release Notes

25 Dec 09:00
028a1bc
Compare
Choose a tag to compare

We're excited to announce the release of llm_onesdk version 0.0.3! This release includes a significant change to the project name and several improvements.

Key Updates:

  1. Project Renamed to llm_onesdk

    • The project has been renamed from 'llm-onesdk' to 'llm_onesdk' for better compatibility with Python naming conventions.
    • Users will need to update their import statements from from llm-onesdk import OneSDK to from llm_onesdk import OneSDK.
  2. Version Upgrade to 0.0.3

    • Incremented version number to reflect recent changes and improvements.
  3. Enhanced Documentation

    • Updated README.md with clearer installation and usage instructions.
    • Improved inline documentation for better code understanding.
  4. Improved GitHub Pages Integration

    • Enhanced rendering of documentation on GitHub Pages.
    • Added a table of contents for easier navigation.
    • Implemented code syntax highlighting for better readability.
  5. Code Optimizations

    • Refined error handling across all supported providers.
    • Improved consistency in API implementations.

Upgrading:
To upgrade to the new version, run:

pip install --upgrade llm_onesdk

Please note that due to the name change, you may need to uninstall the old version first:

pip uninstall llm-onesdk
pip install llm_onesdk

Breaking Changes:

  • The import statement has changed. Update your code to use from llm_onesdk import OneSDK.

We appreciate your continued support and feedback. If you encounter any issues or have suggestions for improvement, please don't hesitate to open an issue on our GitHub repository.

Thank you for using llm_onesdk!

v0.0.2

25 Dec 06:50
Compare
Choose a tag to compare

Full Changelog: https://github.com/LLMPages/onesdk/commits/v0.0.2

Release v0.0.2 (Initial Testing Phase)

We're excited to announce the release of version 0.0.2, marking our first step towards a stable release. This version is now ready for preliminary testing and includes several improvements, particularly in documentation.

What's New

Documentation Enhancements

  • Improved overall content structure for better readability
  • Enhanced clarity of existing documentation
  • Updated examples to reflect current API usage and functionality
  • Fixed typos and grammatical errors throughout

Version Upgrade

  • Bumped version from 0.0.1 to 0.0.2 for initial testing phase

Breaking Changes

While this is a minor version update, please note that this release introduces changes that make it suitable for initial testing. We encourage users to start testing with this version and provide feedback.

Getting Started

To get started with this release, update your dependency to version 0.0.2:

dependency 'llm-onesdk', '0.0.2'

Feedback and Contributions

We welcome your feedback on this initial testing release. Please report any issues or suggestions through our GitHub issues page.

What's Next

We're continuing to work on improving the stability and feature set of our project. Stay tuned for more updates as we move towards our first stable release.

Thank you for your interest and support in our project!