unslothai/unsloth v0.1.904-beta: Train your own Decision model
Turn any text or vision LLM into a Jev-style decision model in Unsloth, with decision accuracy going from 30% to 80%.
ProofCode ↗
Key points
- Train, test, export and serve decision models directly from Unsloth.
- Load ComfyUI diffusion models natively in Unsloth
- Train any text or vision LLM as a Jev-style decision model using QLoRA.
- Use your ComfyUI text encoders and VAEs in Unsloth.
Sources (1)
- [1]unslothai/unsloth v0.1.904-beta: Train your own Decision modelGitHub: unslothai/unsloth · Oct 7, 02:58 PM
Turn any text or vision LLM into a Jev-style decision model in Unsloth, with decision accuracy going from 30% to 80%.
Train, test, export and serve decision models directly from Unsloth.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 6, 2026unslothai/unsloth v0.1.903-beta: New Browser + Voice Cloning
- Oct 6, 2026huggingface/diffusers v0.41.0: Diffusers 0.41.0: QwenImage 2.1 pipeline and more
- Oct 4, 2026ausboss/Qwen-Image-2.1-Outfit-Swap-Consistency-LoRA
- Oct 1, 2026nvidia/PixelUMM
- Oct 1, 2026unslothai/unsloth v0.1.902-beta: Command Palette + Desktop UI/UX
- Sep 29, 2026NVIDIA/TensorRT-LLM v1.3.0rc29