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

[XPU] Enable FP8 training on XPU (#12535)
JoshuaL3000 committed 1 month ago
cb67888
c20595b
Apply the DoRA magnitude in fast_linear_forward (#12642)
Vineeth Sai Varikuntla committed 1 month ago
93434b1
fix: preserve activation QAT in LoRA MLPs (#12558)
cherry77-cloud committed 1 month ago
2db1841