Skip to content

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.

Commits on main

be072ee
Keep weight_decay when embedding_learning_rate is set (#11470)
Vineeth Sai Varikuntla committed 2 months ago
ed004ac
Update _version.py
Daniel Han committed 2 months ago
7cfb922
Qwen-image-2.1
Michael Han committed 2 months ago
77db9ab
feat(chat): make agentic turns survive tab-close (#10365)
CelesteHeartsong committed 2 months ago
0270575
Update pyproject.toml
Daniel Han committed 2 months ago
87026f1
868039c
Update _version.py
Daniel Han committed 2 months ago
7621b25
130fbb6
Avoid 8-bit training optimizer on XPU (#10213)
Ben Younes committed 2 months ago
070f292
6260fcb