v0.1.0-beta

Hey guys, we're super excited to launch Unsloth Studio (Beta), a new open-source web UI to train and run LLMs.

Blog + everything you need to know: https://unsloth.ai/docs/new/studio

* Run models locally on Mac, Windows, Linux
* Compare and battle models side-by-side
* Train 500+ models 2x faster with 70% less VRAM
* Supports GGUF, vision, audio, embedding models
* Self-healing Tool calling / web search + code execution
* Auto-create datasets from PDF, CSV, DOCX
* Export models to GGUF, safetensor and more formats

#### MacOS, Linux, WSL:
For MacOS, ensure you have `cmake` installed. If not, run `brew install cmake`.
```bash
curl -fsSL https://unsloth.ai/install.sh | sh
```
Then to launch every time:
```bash
source unsloth_studio/bin/activate
unsloth studio -H 0.0.0.0 -p 8888
```

#### Windows:
Run in Windows Powershell:
```bash
irm https://unsloth.ai/install.ps1 | iex
```
Then to launch every time:
```bash
.\unsloth_studio\Scripts\activate
unsloth studio -H 0.0.0.0 -p 8888
```

#### Docker
Use our [Docker image](https://hub.docker.com/r/unsloth/unsloth) ```unsloth/unsloth``` container. Run:
```bash
docker run -d -e JUPYTER_PASSWORD="mypassword" \
  -p 8888:8888 -p 8000:8000 -p 2222:22 \
  -v $(pwd)/work:/workspace/work \
  --gpus all \
  unsloth/unsloth
  ```

https://github.com/user-attachments/assets/4f48e6ed-5ef9-42d8-8404-64a4d8b36846

## What's Changed
* Update CODEOWNERS for studio and cli by @danielhanchen in https://github.com/unslothai/unsloth/pull/4266
* [Feature] Support Sequence Classification by @danielhanchen in https://github.com/unslothai/unsloth/pull/4264
* [Feature] VLMs support for GRPO by @danielhanchen in https://github.com/unslothai/unsloth/pull/4265
* [Fix] Respect llm_int8_skip_modules for VLM by @danielhanchen in https://github.com/unslothai/unsloth/pull/4249
* ROCM support by @danielhanchen in https://github.com/unslothai/unsloth/pull/4271
* Remove Blackwell flex attention disable workaround from studio by @danielhanchen in https://github.com/unslothai/unsloth/pull/4273
* ROCM support by @danielhanchen in https://github.com/unslothai/unsloth/pull/4272
* fix: prevent ai-assist model config RCE via untrusted Hugging Face repos by @danielhanchen in https://github.com/unslothai/unsloth/pull/4274
* fix(seed): disable remote code execution in seed inspect dataset loads by @danielhanchen in https://github.com/unslothai/unsloth/pull/4275
* Update CODEOWNERS by @danielhanchen in https://github.com/unslothai/unsloth/pull/4279
* fix: install data-designer plugin non-editable for Colab compatibility by @LeoBorcherding in https://github.com/unslothai/unsloth/pull/4268
* Arch/mixtral by @danielhanchen in https://github.com/unslothai/unsloth/pull/4283
* Improve documentation on how to export model from Colab by @danielhanchen in https://github.com/unslothai/unsloth/pull/4284
* feat: Add Mixtral model support by @danielhanchen in https://github.com/unslothai/unsloth/pull/4285
* Initial changes: Refactor Attention by @danielhanchen in https://github.com/unslothai/unsloth/pull/4286
* patch vlm trainer to resize images by @danielhanchen in https://github.com/unslothai/unsloth/pull/4287
* [WIP] add support for mixtral by @danielhanchen in https://github.com/unslothai/unsloth/pull/4288
* studio: speed up setup -- uv for installs (8x), Ninja for llama.cpp (1.7x) by @danielhanchen in https://github.com/unslothai/unsloth/pull/4289
* fix: remove old comments by @Shine1i in https://github.com/unslothai/unsloth/pull/4292
* PR: Windows Setup Improvements by @rolandtannous in https://github.com/unslothai/unsloth/pull/4299
* miscallenous studio by @Shine1i in https://github.com/unslothai/unsloth/pull/4293
* Fix: Compare Mode Deadlock, Cancel Event Poisoning & IPC Optimization by @rolandtannous in https://github.com/unslothai/unsloth/pull/4303
* studio: fix GGUF inference -- reasoning tokens, max_tokens, server flags, GPU allocation by @danielhanchen in https://github.com/unslothai/unsloth/pull/4290
* chat only with gguf for mac devices by @Manan17 in https://github.com/unslothai/unsloth/pull/4300
* studio: add max steps and epochs toggle switch by @Imagineer99 in https://github.com/unslothai/unsloth/pull/4296
* Fix/colab plugin editable install by @LeoBorcherding in https://github.com/unslothai/unsloth/pull/4281
* Graceful shutdown on Windows (signal handlers for Ctrl+C) by @rolandtannous in https://github.com/unslothai/unsloth/pull/4306
* studio: simplify auth UX to password-only login by @Imagineer99 in https://github.com/unslothai/unsloth/pull/4305
* studio: preserve save_steps when toggling to epochs mode by @Imagineer99 in https://github.com/unslothai/unsloth/pull/4308
* Fix studio frontend build producing empty Tailwind CSS by @danielhanchen in https://github.com/unslothai/unsloth/pull/4311
* Fix setup.sh crash on Mac with empty gitignore array by @danielhanchen in https://github.com/unslothai/unsloth/pull/4313
* [Feature] studio: user can upload eval dataset by @Manan17 in https://github.com/unslothai/unsloth/pull/4307
* fix: Ctrl+C not terminating backend on Linux by @rolandtannous in https://github.com/unslothai/unsloth/pull/4316
* Add download progress bar for non-GGUF models in Chat by @danielhanchen in https://github.com/unslothai/unsloth/pull/4314
* Apply use_reentrant removal to all TRL trainer configs by @danielhanchen in https://github.com/unslothai/unsloth/pull/4321
* Fix VLM GRPO matmul shape mismatch in _get_per_token_logps_and_entropies by @danielhanchen in https://github.com/unslothai/unsloth/pull/4301
* Improve AI Assist: Update default model, model output parsing, logging, and dataset mapping UX by @rolandtannous in https://github.com/unslothai/unsloth/pull/4323
* studio: per-model inference defaults, GGUF slider fix, reasoning toggle by @danielhanchen in https://github.com/unslothai/unsloth/pull/4325
* fix: Resolve CUDA toolkit mismatch on multi-CUDA Windows systems by @rolandtannous in https://github.com/unslothai/unsloth/pull/4324
* [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in https://github.com/unslothai/unsloth/pull/4332
* Fix/colab comment edits by @LeoBorcherding in https://github.com/unslothai/unsloth/pull/4317
* fix: add Qwen3.5 version gate in loader dispatch by @danielhanchen in https://github.com/unslothai/unsloth/pull/4335
* Fix xformers Blackwell guard: broader coverage and root cause docs by @danielhanchen in https://github.com/unslothai/unsloth/pull/4338
* studio: improve Colab notebook, redesign ready popup, and clean up install output by @LeoBorcherding in https://github.com/unslothai/unsloth/pull/4339
* Add check to disable xformers on newer GPUs by @pluesclues in https://github.com/unslothai/unsloth/pull/4342
* studio: training progress, CUDA lib path, dataset_num_proc fix by @danielhanchen in https://github.com/unslothai/unsloth/pull/4336
* studio: fix stale GGUF metadata, update helper model, auth improvements by @danielhanchen in https://github.com/unslothai/unsloth/pull/4346
* studio: show "Off" for repetition penalty = 1 by @danielhanchen in https://github.com/unslothai/unsloth/pull/4349
* studio: update Creative/Precise presets, show "Off" for disabled samplers by @danielhanchen in https://github.com/unslothai/unsloth/pull/4350
* studio: fix slow cancellation of GGUF generation by @danielhanchen in https://github.com/unslothai/unsloth/pull/4352
* Fix: Remove unused `warmupToastShown` variable (TS6133) by @rolandtannous in https://github.com/unslothai/unsloth/pull/4353
* Studio: SVG preview, fix streaming and model selector bugs by @danielhanchen in https://github.com/unslothai/unsloth/pull/4354
* fix: comment out debug print statements by @rolandtannous in https://github.com/unslothai/unsloth/pull/4357
* fix(llm_assist): disable thinking mode for helper model JSON output by @rolandtannous in https://github.com/unslothai/unsloth/pull/4358
* studio: improve onboarding UX, tooltips, and training defaults by @danielhanchen in https://github.com/unslothai/unsloth/pull/4355

## New Contributors
* @LeoBorcherding made their first contribution in https://github.com/unslothai/unsloth/pull/4268
* @Shine1i made their first contribution in https://github.com/unslothai/unsloth/pull/4292
* @Manan17 made their first contribution in https://github.com/unslothai/unsloth/pull/4300
* @Imagineer99 made their first contribution in https://github.com/unslothai/unsloth/pull/4296

**Full Changelog**: https://github.com/unslothai/unsloth/commits/March-2026