v0.1.451-beta
📦 unslothView on GitHub →
✨ 4 features🐛 33 fixes🔧 6 symbols
Summary
This release focuses heavily on bug fixes and cross-platform support improvements, particularly within Unsloth Studio, and updates the installation mechanism to ensure users receive the latest package versions.
Migration Steps
- To update Unsloth or install a new Unsloth Studio, use the provided platform-specific installation scripts instead of `unsloth studio update`.
- macOS, Linux, WSL users should run: `curl -fsSL https://unsloth.ai/install.sh | sh`
- Windows users should run: `irm https://unsloth.ai/install.ps1 | iex`
✨ New Features
- Studio: enable audio input for Gemma 4 GGUFs
- Studio: enable GGUF tools with vision inputs
- feat(studio): Hub + Download Manager
- feat: support text-only loading of Gemma 3 27B via FastLanguageModel (skip SiglipVisionModel)
🐛 Bug Fixes
- Port KTO logps truncation guard to TRL 1.x _compute_logps refactor (later restored)
- Studio: disable mlx gc for none
- Fix chat text cutoff at composer dock and speed up plus icon spin
- Studio: refine tool call and reasoning trigger UI
- fix: warn when localhost resolves to ::1 but Studio is bound only to 127.0.0.1
- fix: persist Studio thread synchronously on first runStart so mid-stream refresh keeps the prompt
- Studio: accept system-role messages in Claude Code requests
- Studio: fix load_freeze audio-type tests for #6000's Gemma 4 <|audio|> probe
- Studio: fix chat preset persistence with fast mode
- Studio: fix Repo tests (CPU) by stopping the ROCm test from leaking a fake utils into sys.modules
- Studio: stop ROCm amd-smi tests leaking a fake loggers into sys.modules (follow-up to #6027)
- Studio: tag MLX loaded models as MLX instead of Base in chat
- Studio: remove red border on chat error messages
- Studio: keep chat in place when composer attachments resize it
- Studio: stop leaking internal exceptions to API clients; harden sandbox path
- Fix MoE LoRA target parameter handling
- qwen 3.5 export fixes
- [fix] Nvfp4 load
- Studio fix recipe dataset preview
- Studio: improve OpenAI- and Anthropic-compatible API spec compliance
- Studio: follow-up fix for GGUF developer prompts
- Studio: stop the providers dialog from resetting custom provider config state
- Studio: clean-room compact RAG (knowledge bases, hybrid search, fast indexing)
- studio(ui): use the --primary brand token for the avatar fallback color
- fix(studio): block arbitrary external image URLs in markdown renderer
- Studio: center account avatar vertically in sidebar footer pill
- fix: clearer Studio setup error when GPU driver is too old for the installed CUDA toolkit
- Studio: faithful conversation export and import round trips (ShareGPT system role, CSV quoted newlines)
- Windows/WSL installer: fix winget msstore cert failure, amd-smi DiskPart prompt, and enable AMD GPU (Strix Halo gfx1151)
- Studio: fix nested dropdown submenus clipped by the menu alignment nudge
- fix/validate dataset video paths before training
- Restore config use_cache in for_inference after gradient checkpointing prep
- fix: persist Windows ROCm BNB version