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
Studio: pick which LoRA to add from a searchable list instead of cycling through them (#16964)
671a16c oobabooga committed 13 days ago
Studio: report the LTX-2.5 encoder worker host memory on macOS and Windows too (#17014)
66b6062 Daniel Han committed 13 days ago
Studio: keep Hermes MCP servers in unsloth start hermes (#16949)
4030f3f Nilay committed 13 days ago
Studio: accept zstd-compressed requests from the TypeSafe SDK (#16929)
5252459 Nilay committed 13 days ago
Studio: keep same-second fine-tuning jobs in submission order across restarts (#16835)
4ee2a51 oobabooga committed 13 days ago
Studio: key the modal pointer-lock rules off attributes that never sit above the thread (#17006)
be2e825 Daniel Han committed 13 days ago
Studio: keep the base URL query string on connection requests (#16951)
7256cb4 Nilay committed 13 days ago
Studio: reload an idle-unloaded GGUF with its original launch settings (#16927)
847a8ec Nilay committed 13 days ago
CI: key the Playwright browser caches on the OS release, and move the WebKit jobs back to ubuntu-latest (#17009)
c21cc33 Daniel Han committed 13 days ago
Studio: keep Codex AGENTS.md, skills and rules in unsloth start codex (#16947)
e59af46 Nilay committed 13 days ago
Studio: stop cutting Claude answers at 1,024 tokens without a limit (#16953)
183f507 Nilay committed 13 days ago
Studio: cap chat code blocks at a scrollable height and keep them coloured while scrolling (#16865)
7e22973 Michael Han committed 13 days ago
Studio: auto-start the server from the same install as unsloth start (#16948)
6106dfc Nilay committed 13 days ago
Studio CI: guard the long-chat gate Glass ceilings against runner noise (#17012)
74b0cbe Daniel Han committed 13 days ago
Studio: stop Export all chats from saving an empty or stale backup (#16930)
7e0143e Nilay committed 13 days ago
Studio: Agents tab says unsloth start claude needs a GGUF model (#16952)
729320c Nilay committed 13 days ago
Studio: offer ChatGPT Reconnect for reused or revoked refresh tokens (#16946)
d94eccf Nilay committed 13 days ago
Studio: block more metadata host spellings for approved Python (#16937)
58a1443 Nilay committed 13 days ago
Studio: size the popular audio.cpp TTS and music models from measured peaks on each runtime backend (#16984)
115fbb2 oobabooga committed 13 days ago
Studio: keep messages from another window when deleting in a stale one (#16926)
634b4f4 Nilay committed 13 days ago
Studio: stop OpenAI chats failing on earlier replies sent as parts (#16945)
25d4692 Nilay committed 13 days ago
Studio: let unsloth start attach to a loaded Ollama model (#16944)
66dfa6d Nilay committed 13 days ago
Studio: start a first video load's cache writes after its first render (#17001)
482514b Daniel Han committed 13 days ago
Studio: quantise the hosted int8 leftovers on the GPU on offloaded loads (#17002)
5505f7e Daniel Han committed 13 days ago
Studio CI: long-chat gate Glass ceilings back to 10x (areas) and 18x (Settings scroll) (#17003)
8e8c910 Daniel Han committed 13 days ago
Studio images: render on the NVIDIA card when torch drives no GPU (#16853)
5a110a6 Debojyoti Banerjee committed 13 days ago
Studio RAG: embed the documents of one ingest together instead of one call per document (#16727)
a82cd33 Debojyoti Banerjee committed 13 days ago