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: inline message actions, matching response editor (#13615)
274b534 Michael Han committed 1 month ago
tests: resolve every closure name in the restoreSettings sandbox (#13617)
85c2c2f Daniel Han committed 1 month ago
Studio: Auto drops serving slots before the MTP drafter (#13455)
b93d4d1 Daniel Han committed 1 month ago
Fix the diffusion scheduler, parity, formatter and bundle reds on main (#13612)
b8d76e1 Daniel Han committed 1 month ago
Studio: list, reveal and delete audio.cpp umbrella models by folder (#13519)
3207f04 Etherl committed 1 month ago
FastSentenceTransformer: reuse the loaded model for Transformer subclass modules (#13591)
2e556ba Daniel Han committed 1 month ago
Studio: LTX-2.5 GGUF quants (unsloth/LTX-2.5-GGUF) (#13573)
c35052e Daniel Han committed 1 month ago
Studio: fit MiniMax-H3 GGUF to a Vulkan host's GTT pool before rendering (#13531)
cd8b2ed Daniel Han committed 1 month ago
Studio: optional Chart view with labelled axes for weekly and cumulative token activity (#13558)
8f9ea7a Daniel Han committed 1 month ago
Studio: faster browser agent page reads after a load, and leaner drags (#13602)
dd5ec48 Daniel Han committed 1 month ago
Studio: fix audio.cpp partial downloads, aligner progress, CPU fallback (#13515)
00b323c Etherl committed 1 month ago
Studio: video component sizes from safetensors headers, measured peaks for every video family (#13537)
bc3114d Daniel Han committed 1 month ago
Keep Mistral Small 3.2 tool definitions in training rows (#13597)
44b55b8 Daniel Han committed 1 month ago
Studio: keep the prompt cache across long tool chats and follow-up turns (#13456)
fd30659 Daniel Han committed 1 month ago
Studio: add /compact and draw compactions where they happened (#13562)
a062c04 Daniel Han committed 1 month ago
Studio: fix Spark-TTS training failing on None/config.yaml (BiCodec repo root unset) (#13465)
7c2b8e9 CarlitoGwapito committed 1 month ago
Studio: keep the Transcribe loaded state and Eject in step with the sidecar (#13514)
6aaa1f3 Etherl committed 1 month ago
Studio: add a Download to folder for model downloads (#13557)
280c456 Daniel Han committed 1 month ago
Studio: pick the sampler on the Images page and load v-prediction SDXL checkpoints correctly (#13569)
d1d7495 Daniel Han committed 1 month ago
Studio: let coding agents use Unsloth Studio over MCP with API keys (#13118)
aa7f76d Etherl committed 1 month ago
Tests: follow the sidebar rename dialog and run the H3 compile-order test without diffusers (#13590)
e033030 Daniel Han committed 1 month ago
Studio: apply the tool output length setting to MCP tool results (#13588)
cb4c624 Daniel Han committed 1 month ago
Studio: load sidebar chats, projects and training runs 20 at a time from the server (#13420)
6db2bf0 Daniel Han committed 1 month ago
Studio setup: build the frontend in the background on source installs (setup.sh and setup.ps1) (#13442)
f5a45dd Daniel Han committed 1 month ago
studiobench UI parity: report a UI change the PR declares instead of failing (#13583)
481c979 Daniel Han committed 1 month ago
Make the weights_at_root call-site and diffusion autoopt parity tests host independent (#13582)
2a4d4df Daniel Han committed 1 month ago
Run non-LoRA PEFT adapters through their own forward in fast inference (#13578)
51d8ca3 Daniel Han committed 1 month ago
Studio: downscale oversized image uploads and add crop to image inputs (#13561)
7b3cb9f Daniel Han committed 1 month ago