Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.
Python agentaichatgptdeepseekfine-tuninggemmaimage-generationllamallmllmsopenaipythonqwenreinforcement-learningself-hostedstable-diffusiontext-to-speechttsuiunsloth
Commits on main
docker: give unsloth/unsloth-rocm a Hub page and keep it in sync (#11059)
c5faa7f Daniel Han committed 2 months ago
Studio: fix structured columns in Data Recipes on non-GGUF models (#11049)
9a127b6 Nilay committed 2 months ago
Studio: keep waiting on a llama-server load that is still reading its model (#11055)
98f15da oobabooga committed 2 months ago
Assert the preset sheet subscribes, not how the subscription is spelled (#11065)
5d98906 Daniel Han committed 2 months ago
Draw the probe port from below the ephemeral range (#11086)
ae6bc9e Daniel Han committed 2 months ago
Give the launcher signal tests a worker to themselves (#11085)
9d72fad Daniel Han committed 2 months ago
Answer /status the runtime fields a backend double cannot invent (#11076)
b5f26f6 Daniel Han committed 2 months ago
Docker publish: verify the Studio image env, and stop telling latest users to opt in to CPU (#11071)
87cb503 Daniel Han committed 2 months ago
Windows installer: stop uv splitting a requirements path, and keep the error position in diagnostics (#11064)
0404030 Daniel Han committed 2 months ago
Release: sign every shipped PowerShell script and publish digests for the rest (#11069)
6731ba1 Daniel Han committed 2 months ago
Stop the deps CLI fixture racing another worker for the requirements tree (#11063)
71d85d5 Daniel Han committed 2 months ago
README: mount the Hugging Face cache in the Docker quickstart (#11062)
2ab07c9 Daniel Han committed 2 months ago
docs: clarify the Docker quickstart shell, Windows setup and logs command (#11036)
7e8c8fa Lee Jackson committed 2 months ago
Recognise a partially initialised torchvision and fix the studiobench ab_plan row (#11015)
764a0e4 Daniel Han committed 2 months ago
Unsloth Studio Installer: improve GPU print details (NVIDIA, Intel, AMD) (#8762)
32c01c8 Leo Borcherding committed 2 months ago
Studio: re-measure the startup bundle budget, which main is red on by itself (#11013)
a58537e Daniel Han committed 2 months ago
fix(amd:audio): decode audio through PyAV when torchcodec installs but cannot load (#8642)
fe90d0d Leo Borcherding committed 2 months ago
Studio backend tests: settle the real loggers package in conftest for repo-root pytest runs (#11028)
48b0831 jishnav committed 2 months ago
studio: add prompt queue controls and composer preferences (#10980)
f4958ce Michael Han committed 2 months ago
Studio: accept OpenAI video_url content parts on /v1/chat/completions (#10439)
f7ab209 Long Yixing committed 2 months ago
Measure whether a hardening change alters what the installer does, instead of arguing that it does not (#10992)
6370afe Daniel Han committed 2 months ago
studio: organize data settings and make archives searchable (#10974)
0ab40fe Michael Han committed 2 months ago
Ask AMSI whether the installer compiles, which is the mechanism #10805 actually reports (#10994)
89862da Daniel Han committed 2 months ago
Delete the emit apparatus that existed to colour a banner, now that measurement says it was a no-op (#10999)
4becbbd Daniel Han committed 2 months ago
fix(studio): keep recipe workers alive when request threads retire (#11021)
73eb6fb alkinun committed 2 months ago
WSL installer: stop running the C# compiler on the user's Windows machine (#10978)
201790e Daniel Han committed 2 months ago
Studio: refuse tool access to the auth directory and mask keys in tool output (#10895)
56d1946 Daniel Han committed 2 months ago