AionUi v1.2.2 Official Release Notes

This is the official, stable release of AionUi v1.2, following the previous v1.2.0-preview.
We strongly recommend all users upgrade to v1.2.2 for the most stable and feature-rich experience. This version fully delivers on the vision of the 1.2 series. If you have any questions or encounter issues, please let us know!

🚀 Highlight Features in the 1.2 Series

1. API Key Auto-Rotation

• Easily configure and use multiple API keys for your models. The system automatically rotates keys and temporarily blacklists problematic ones to minimize interruptions and avoid rate limits.
• Allow the modification of the LLM platform name and support diverse configurations for one platform

2. Multi-Agent Collaboration

AionUi now supports multi-agent mode, enabling collaboration between different LLMs—including Claude Code and Qwen Code. Easily switch and coordinate between agents for more complex workflows.

3. Advanced Image Generation Tools

• Enhanced, user-friendly settings for image generation.
• Support for generating multiple images at once.
• Improved configuration guides and documentation links for a smoother setup.

4. Robust Tool Configuration & Compatibility

• More flexible tool configuration system.
• Enhanced type system and backward compatibility.
• Automatic migration and rollback for key configuration items.

5. User Experience & Performance Improvements

• Redesigned welcome page and input styling.
• Chat history recovery for better continuity.
• Major codebase refactoring for improved type safety and debugging performance.

6. ACP Architecture Upgrade (v1.2.2)

• Upgraded the core AI Communication Protocol (ACP) from 0.1.6 to 0.4.0 for more stable and compatible connections.
• Refactored the type system for higher runtime safety.
• Simplified connection configuration and improved error/message handling.
• Cleaner codebase and improved dependency management.

What's Changed

• feat: update docs v1.2.0 with multi-agent mode and API key rota… by @IceyLiu in #135
• feat: enhance image generation and custom model support by @goworm in #138
• feat: enhance Gemini model list support with dynamic API fetching by @goworm in #140
• feat: upgrade to version 1.2.2 with ACP architecture improvements and bug fixes by @kaizhou-lab in #143

Full Changelog: v1.2.0...v1.2.2