v3.4.0 v3.5.0
xorbitsai/inferenceSep 11, 2026 → Sep 25, 202614 days between releases
What's New
Adds support for new audio, image, and LLM models while introducing MLX optimizations and observability.
New Features
- Spark-X2.5 LLM support and exposed native concurrency.
- Support for new audio models (YuE2, AuK, Irodori) and MLX.
- Opt-in model request logging with source identity.
- LingBot-World-V2 support and MinerU2.5 model support.
- Enhanced TTS support (Irodori v4.1, Breeze 2) with MLX.
Improvements
- MLX optimizations for audio and TTS models.
- vLLM batching enhancements for performance.
- Propagation of correlation metadata across actor calls.
- Native concurrency exposure for image/video engines.
- Comprehensive logging and observability capabilities.
Upgrade Notes
Ensure your environment meets the xoscar 0.10.0 and Python 3.14 requirements. Configure the new opt-in model request logging feature in your settings for observability.
The Two Releases
v3.4.0
Sep 11, 2026· 1 month ago
This release introduces support for new models such as Gemma-4 and MiniCPM5-2B, alongside resumable cache management and multi-worker scaling. It also adds OCR capabilities, OCI URI support, and significant Web UI enhancements including TTS streaming and new language options.
Full release notesv3.5.0
Sep 25, 2026· 26 days ago
This release addresses a security vulnerability regarding custom-model registration and introduces extensive new features including support for various audio, image, and LLM models, along with logging and monitoring capabilities.
Full release notes