Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.8, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM5.3, Gemma4, Llava, Phi4, ...) (AAAI 2025).
LMDeploy is a toolkit for compressing, deploying, and serving LLMs.
A Next-Generation Training Engine Built for Ultra-Large MoE Models
A lightweight framework for building LLM-based agents
Fast multi-modal decision model
InternLumina-U2: A Multi-Codebook Diffusion Large Language Model for Omni-Visual Understanding, Image Generation and Editing
ResOPD: Tail Residualization for Sparse On-Policy Distillation. Core implementation built on verl.
Intern-S2-Mobius
An official implementation of "SpatialASM: Internalizing Active Mentalization for Spatial Reasoning"