unsloth

Unsloth lets you run, train, and deploy AI models locally, with support for all types of models including Kimi K3, MiniMax-H3, Qwen3.8, Muse Glimmer, DeepSeek-V4, Gemma 4. It supports reinforcement learning, LoRA, QLoRA, full fine tuning, pretraining, RL, GRPO, DPO, and FP8.