[CVPR 2025] Magma: A Foundation Model for Multimodal AI Agents

Code for the Molmo Vision-Language Model

Mixture-of-Experts for Large Vision-Language Models

[CVPR2025 Highlight] Insight-V: Exploring Long-Chain Visual Reasoning with Multimodal Large Language Models

AllenAI's post-training codebase

Official inference framework for 1-bit LLMs

VPTQ, A Flexible and Extreme low-bit quantization algorithm

Qwen2.5 is the large language model series developed by Qwen team, Alibaba Cloud.

SEED-Voken: A Series of Powerful Visual Tokenizers

VMamba: Visual State Space Models，code is based on mamba

Implementation for "Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs"

A high-throughput and memory-efficient inference and serving engine for LLMs

Qwen2.5-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

Video datasets

Distributed Training Over-The-Internet

Beyond Straight-Through

VILA is a family of state-of-the-art vision language models (VLMs) for diverse multimodal AI tasks across the edge, data center, and cloud.

✨✨VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interaction

For optimization algorithm research and development.

An Easy-to-use, Scalable and High-performance RLHF Framework (70B+ PPO Full Tuning & Iterative DPO & LoRA & RingAttention & RFT)

[NeurIPS 2024] An official implementation of ShareGPT4Video: Improving Video Understanding and Generation with Better Captions

SALMONN: Speech Audio Language Music Open Neural Network

AI2-THOR Data Collection Tool Based On Keyboard Interaction

OpenVLA: An open-source vision-language-action model for robotic manipulation.

A flexible and efficient codebase for training visually-conditioned language models (VLMs)

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

Code for the Paper M4U: Evaluating Multilingual Understanding and Reasoning for Large Multimodal Models.

DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model