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
Fix Hub On Device Selection Across Refresh and Download Changes (#9922)
3574e20 Eyera committed 2 months ago
Show activity on a 0-byte Xet download and keep finished cards visible longer (#9303)
23794ec Nilay committed 2 months ago
Studio: fix clipped settings scrollbar on Windows WebView2 (#10074)
2da3771 Nilay committed 2 months ago
Studio: resume HTTP after Xet stall without a transport-conflict banner (#10045)
c47645f Souravrajvi0 committed 2 months ago
CI: fix llama.cpp seed caching, Diffusers wheel installs, and isolated test coverage (#9979)
1400031 oobabooga committed 2 months ago
Say why Studio could not enumerate a GPU, instead of just logging [] (#10056)
489e85a Daniel Han committed 2 months ago
Studio: notice a quarantined payload before trusting the update fast path (#10053)
fa57a79 Daniel Han committed 2 months ago
fix(studio): read tool-call arguments as the types their schema declares (#10019)
e5767a0 Long Yixing committed 2 months ago
fix(studio): normalize Muse Glimmer's recipient-addressed protocol (#8422)
10fe271 Long Yixing committed 2 months ago
Studio: install torch from an index that has kernels for the user's AMD GPU (#9829)
38440ee oobabooga committed 2 months ago
Read the chat storage keys on call, not at module scope (#9844)
f7b084a Daniel Han committed 2 months ago
Studio: keep a Windows update from leaving a CPU-only PyTorch (#9857)
d5b153c Daniel Han committed 2 months ago
Stop shipping test suites in the wheel and the sdist (#10060)
6179e04 Daniel Han committed 2 months ago
Fix excessive studio.db writes during chat inference (#9945)
eb2c6de Lee Jackson committed 2 months ago
Let tool chats decode in parallel instead of one at a time (#10046)
044cdb8 Daniel Han committed 2 months ago
Use the head-aware planner for Studio's multi-GPU device map (#10042)
eab5cda Daniel Han committed 2 months ago
Wait for the idle batch flush instead of racing a fixed sleep (#10063)
8fba17f Daniel Han committed 2 months ago
Studio: set `GGML_CUDA_ENABLE_UNIFIED_MEMORY` only where it gains memory (#9884)
16d596d oobabooga committed 2 months ago
feat(export): GGUF shard size control, save directory override, fix private Hub push (#6107)
8f91a3f Leo Borcherding committed 2 months ago
Stop the GGUF header and the chat detail polls from filling the log (#10040)
256db97 Daniel Han committed 2 months ago
Tag the compat endpoint's Ollama rows with their real source (#9988)
d37e98e Shubhankar Tripathy committed 2 months ago
Use the plain download icon in the Hub On Device views (#10055)
6b60cb8 Michael Han committed 2 months ago
unsloth start: stop --allowedTools swallowing the forwarded prompt (#9939)
ca37900 Nilay committed 2 months ago
Stop the llama-server probe deny-list answering 405 (#10049)
57ef43c Daniel Han committed 2 months ago
Answer llama-server discovery probes instead of serving them the app shell (#10037)
640f0db Daniel Han committed 2 months ago