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
Prepare an exhausted call before the budget gate, so its replay parses (#10274)
a61d639 Shubhankar Tripathy committed 1 month ago
tests(version_compat): AST-based symbol checks and per-model drift coverage (#6740)
a1ff98b Suchitra Malimbada committed 1 month ago
Chat: show the Thinking control for Ollama models that report the thinking capability (#10043)
afd6a6c Shubhankar Tripathy committed 1 month ago
Studio: recount tokens when status sync adopts a resident GGUF (#10467)
b921a2c Chaitanya Laxman committed 1 month ago
Revert "Studio: add per-model custom llama.cpp INI configuration (#10783)" (#12725)
dd3759c Daniel Han committed 1 month ago
Compiled decode for gpt-oss when unsloth_zoo has the routed experts patch (#12713)
1a073ea Daniel Han committed 1 month ago
fix(template): remove duplicate frontmatter block from bug issue template (#10392)
0b2b0b2 陈志谦 committed 1 month ago
Studio: serialize studio.db connection opens and closes (#10069)
86e7068 Kenshi committed 1 month ago
Studio: list a model once when a custom folder overlaps LM Studio (#9169)
53e7795 Kenshi committed 1 month ago
Studio: let an attachment use Queue while a reply is running (#9215)
988ab87 Kenshi committed 1 month ago
Studio: panel-left / panel-right icons for open sidebars, API and home-wifi icons (#12720)
4142ec3 Michael Han committed 1 month ago
Studio: fade Run settings at the edges it scrolls past (#12722)
a9c7a3d Michael Han committed 1 month ago
Studio: keep FlashAttention 4 inside a fullgraph compile (#12686)
5ac49d4 Daniel Han committed 1 month ago
fix: preserve logits in fast cross entropy backward (#12351)
0cf426b cherry77-cloud committed 1 month ago
fix: preserve shared upstream gradients in RoPE backward (#12528)
d81de9f cherry77-cloud committed 1 month ago
Studio: request stream usage from llama.cpp connections so the context bar fills (#12688)
fb889c2 Jay Zhou committed 1 month ago
Bump Desktop crates, Docker JupyterLab, setuptools and CI pins for advisories (#12704)
57937d2 Daniel Han committed 1 month ago
Studio: add a Convert audio page for voice conversion with A/B compare (#12611)
4444657 Etherl committed 1 month ago
Studio: add the Separate audio workflow with a synced stem mixer (#12610)
4c09908 Etherl committed 1 month ago
MLX inference test stub: accept any arguments a zoo helper is called with (#12700)
2d5863c Daniel Han committed 1 month ago
Studio: add the Music studio with song, sound effect and edit modes (#12609)
c601438 Etherl committed 1 month ago
Studio: add timestamps, speakers and exports to the Transcribe page (#12607)
9d76f75 Etherl committed 1 month ago
Studio: add the Clone page, audio inputs and saved voices (#12602)
846989d Etherl committed 1 month ago
Studio: split Audio into Speak, Music and Transcribe workspaces (#12600)
c754470 Etherl committed 1 month ago