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: do not permanently purge RAG when a project is recreated (#10583)
1e212fe Apurv Gupta committed 1 month ago
Studio: opt-in recursive scanning for custom model folders (#6989)
201c28f Aditya Aggarwal committed 1 month ago
Studio: keep unfinished chat links as text instead of [blocked] (#12104)
7aded1f UmranPros committed 1 month ago
studio: report a denied rename as a possible ACL fault, not a scanner (#10075)
2dc042c Wayne Sletcher committed 1 month ago
Studio: seam-free VAE tiles for Qwen-Image-2.1 on low-VRAM tiers (fixes thin horizontal / vertical lines) (#12696)
3161553 Daniel Han committed 1 month ago
Studio: hide iq4_nl KV cache where llama.cpp runs its attention on the CPU (#7008)
4ccad39 Rod Boev committed 1 month ago
Studio: int8 activations on T4 for Qwen-Image (2.5x per step) (#12690)
7fce87b Daniel Han committed 1 month ago
Offloaded embedding: one graph under compiled inference (#12726)
1ac416b Daniel Han committed 1 month ago
Studio: return a missing number in a recipe dataset page as null (#12103)
72a5990 Mohammad Hijjawi committed 1 month ago
Studio: show a JSON seed's values as written in the Data Recipe preview (#12180)
bdd3b00 Mohammad Hijjawi committed 1 month ago
CLI: show a thinking model's reasoning with --think when attached to a running Unsloth (#11956)
06bfeb9 breken committed 1 month ago
studio: report why launcher recovery failed, not that it was absent (#9924)
3fd29fd Wayne Sletcher committed 1 month ago
Studio: resolve GET /v1/models/<id>:<quant> for an on-disk quant (#12727)
3cef25c Daniel Han committed 1 month ago
Studio: bound the code highlighter cache for tool cells, previews and READMEs (#10609)
8254d6e Chaitanya Laxman committed 1 month ago
Studio: open a pasted-text upload without its <pasted_text> wrapper (#12198)
3de9fce Mohammad Hijjawi committed 1 month ago
Studio: screen reader accessibility improvements (NVDA/JAWS) (#6601)
df6fd32 devinprater committed 1 month ago
Studio: show the models drive in the Live Monitor when it is not the system disk (#12724)
aec4558 Daniel Han committed 1 month ago
Add Intel GPU (XPU/SYCL) auto-detection to the Linux installer (#9084)
617b0d4 Jan committed 1 month ago
tests: skip two version_compat suites where the daily sweep has no torch (#12274)
735b9e4 Wayne Sletcher committed 1 month ago
Studio: find a hand-added mmproj when the GGUF repo publishes none (#9365)
874fb67 yzxcj797 committed 1 month ago
Studio browser: hide tab menu items that don't apply, use the toolbar's reload icon (#12728)
bb9f2db Michael Han committed 1 month ago
Studio: bake a relocatable RUNPATH into the Linux llama.cpp source build (#12398)
d97fe69 Shubhankar Tripathy committed 1 month ago
Studio: stop link definition lookalikes forcing full-document streaming renders (#10688)
df4eb29 Apurv Gupta committed 1 month ago
Studio: separate active context from processed tokens (#9528)
801fa7c dre4moff committed 1 month ago