All releases
Auxot OSS v0.1.9 February 27, 2026
Vision, Image Generation, and Model Registry 1.1.2
Added multimodal vision and image generation support, upgraded the model registry to 1.1.2 with Qwen3.5-35B-A3B as the new default model.
Features
- Added vision support via
mmprojmultimodal projectors, enabling workers to process and reason over image inputs. - Added image generation capabilities, allowing agents to produce images as part of task outputs.
Improvements
- Upgraded the bundled llama runtime to b8157 and model registry to 1.1.2.
- Changed the default model to Qwen3.5-35B-A3B for improved performance and tool-use quality.
- Updated tools worker to align with new model and runtime capabilities.