🚀 Xinference v2.2.0 Release Notes

        ✅ Highlights
        🧠 Next-Gen LLM Support
                GLM-5
Kimi-K2.5
MiniMax-M2.5
Qwen3.5

            

    

🌐 Community Edition Updates

📦 Installation

Via pip: pip install 'xinference==2.2.0'

Via Docker: Pull the latest image, or update using pip inside the container.

🆕 New Model Support

GLM-5
Kimi-K2.5
MiniMax-M2.5
Qwen3.5

✨ New Features

Supported GLM-5 and Kimi-K2.5 running on the vLLM engine.
Updated related model configurations.

🐞 Bug Fixes

Fixed the multi-file processing issue in create_image_edits.
Replaced 55 bare except clauses to improve exception handling standardization.

📚 Documentation Updates

Updated v2.1.0 documentation.
Added Docker pull instructions to the README.

🏢 Enterprise Edition Updates

🔧 Enhanced PPU Support
Optimized execution and scheduling capabilities in PPU environments, improving the stability and performance of enterprise-level deployments.

🤖 XAgent v0.1.2

Added PPT generation capabilities, and optimized frontend user experience along with overall stability.