Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Gemma4, Llava, Phi4, ...) (AAAI 2025).
Introduction
魔搭模型社区的大模型训练与推理工具箱,支持LLaMA、千问、ChatGLM、百川等多种先进模型,以及LoRA、ResTuning、NEFTune等多种训练方法。【此简介由AI生成】
Apache-2.0 Python3.49 KCommitsdeepseek-r1embeddinggrpointernvlligerllamallama4llmmegatronmoemultimodalopen-r1peftqwen3qwen3-6qwen3-omniqwen3-vlrerankersft
Customize your domain