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
Accept a multimodal processor in get_chat_template and construct_chat_template (#10152)
ba7a539 Aditya Malik committed 2 months ago
Fix Studio preferring MTP sidecars over embedded heads (#10181)
368e1cb Wasim Yousef Said committed 2 months ago
Test the non-GGUF status contract by behaviour, not by source spelling (#10185)
ecbe3dd Daniel Han committed 2 months ago
Stop a new import in a module under test breaking the tests that load it (#10186)
e84ef62 Daniel Han committed 2 months ago
Signal the research cancellation tests instead of sleeping a guessed interval (#10190)
d45704f Daniel Han committed 2 months ago
Security audit: re-approve fastmcp-slim 4.0.x and unsloth-zoo 2026.8.17 (#10187)
ca161bd Daniel Han committed 2 months ago
Update CODEOWNERS for current maintainers and code areas (#10191)
9461e3a Daniel Han committed 2 months ago
Trim comments in studio/backend routes, utils and installers (#10112)
75986a6 Daniel Han committed 2 months ago
Trim comments in the studiobench harness and unsloth_cli (#10118)
25c10a3 Daniel Han committed 2 months ago
Trim comments in studio/backend core services and the unsloth package (#10116)
115339b Daniel Han committed 2 months ago
Reflow comments in studio/backend/core/inference to 120 columns (#10115)
f580f73 Daniel Han committed 2 months ago
Studio: test that a chat message keeps its identity (#9984) (#10064)
9ea157e UmranPros committed 2 months ago
Studio: stop Deep Research from throwing away a finished report (#10166)
7638383 Nilay committed 2 months ago
Stop the kwarg-spacing post-pass dying on Python 3.11, which reformatted the repo (#10137)
c1ec626 Daniel Han committed 2 months ago
Studio: keep both branches when a chat's first message is edited (#10163)
381b9fc Nilay committed 2 months ago
Studio: keep tool cards in place when you edit a reply (#10162)
2694d7e Nilay committed 2 months ago
Studio: don't size attached documents by a local model on hosted chats (#10164)
7ae1304 Nilay committed 2 months ago
Studio: don't re-prompt after model produced a complete answer (#5714)
b648158 Daniel Han committed 2 months ago
Studio: honor a cleared or disabled speculative decoding setting (#5696)
b3c4010 Daniel Han committed 2 months ago
Studio: dispatch on OpenAI agentic web_search action variants (#5787)
7504feb Daniel Han committed 2 months ago
Fix llama.cpp Studio update failures on GitHub API rate limits (#9970) (#9980)
1a7e11b Souravrajvi0 committed 2 months ago
Studio: surface every live provider model in the picker (#5684)
5c8c238 Daniel Han committed 2 months ago
Studio: restore sticky checkpoint boundaries on tool-heavy threads (#9971)
193841e alkinun committed 2 months ago
fix(studio): poll inference status while CLI model is loading (#5901)
56b9ca4 James Dawdy committed 2 months ago