Morph MORPH
SIGN IN SIGN UP
unslothai / unsloth UNCLAIMED

Unsloth Studio is a web UI for training and running open models like Qwen, DeepSeek, gpt-oss and Gemma locally.

0 0 98 Python

ACTIVITY

Commits Reviews
week month year
COMMITS
20
in the last year
CONTRIBUTORS
5
active
STARS
0
total
FORKS
0
total
TOP CONTRIBUTORS
D
Daniel Han
12 commits
O
oobabooga
3 commits
L
Lee Jackson
3 commits
W
Wasim Yousef Said
1 commit
L
Leo Borcherding
1 commit
RECENT COMMITS
O
Studio: optimize chat streaming by batching renders to one per animation frame (#5788)
oobabooga
de21723
L
Studio: polish model load toast styling (#5648)
Lee Jackson
a05944a
O
studio/frontend: pad Python tool code block to fix corner clipping (#5938)
oobabooga
07f0edd
W
fix(chat): refine artifact panel styling (#5922)
Wasim Yousef Said
aee0657
L
Studio: manage chat history with projects (#5725)
Lee Jackson
e0ff6a1
L
Studio: support connected models in compare mode (#5824)
Lee Jackson
2186577
D
Bump install.sh / install.ps1 pin to unsloth>=2026.5.10 (#5931)
Daniel Han
e94490e
O
Studio: clearer MCP server validation when stdio is disabled (#5928)
oobabooga
0cbf641
D
Studio: move MCP to a composer button with presets (#5926)
Daniel Han
a0b85bc
D
Studio: cover B300 (sm_103) with the Linux prebuilt bundles (#5930)
Daniel Han
ae5d607
D
Update _utils.py
Daniel Han
0296984
L
Studio: forward --has-rocm from setup.sh when gfx resolution fails (#5927)
Leo Borcherding
f1a0581
D
Studio CI: tolerate transient runner crashes in the llama-server log collection step (#5925)
Daniel Han
52612d6
D
Studio: forward the resolved AMD gfx arch to the prebuilt installer (#5923)
Daniel Han
86b1918
D
Studio CI: run the Resolve-CudaToolkit unit test in the Windows GGUF job (#5921)
Daniel Han
ef43b5f
D
Studio: source-build arm64 Linux GPU hosts, with a CPU prebuilt fallback (#5924)
Daniel Han
172d9d1
D
Studio: extend the pinned Blackwell GPU fallback to CUDA 13.0 drivers (#5920)
Daniel Han
f695fbd
D
Studio: match the Linux llama.cpp prebuilt to the runtime cudart major (#5914)
Daniel Han
ab7a7f9
D
Studio CI: tolerate transient artifact-upload flakes on diagnostic log steps (#5913)
Daniel Han
0091b72
D
Studio: defer the Windows CUDA Toolkit check so prebuilt users are not blocked (#5912)
Daniel Han
fc09d8f