MiniCPM-V 4.5: A GPT-4o Level MLLM for Single Image, Multi Image and High-FPS Video Understanding on Your Phone

Official implementation of BLIP3o-Series

VeOmni: Scaling Any Modality Model Training with Model-Centric Distributed Recipe Zoo

Contexts Optical Compression

Fully Open Framework for Democratized Multimodal Training

StreamingVLM: Real-Time Understanding for Infinite Video Streams

Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"

Official codebase used to develop Vision Transformer, SigLIP, MLP-Mixer, LiT and more.

20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.

Minimalistic large language model 3D-parallelism training

Training Large Language Model to Reason in a Continuous Latent Space

Official codebase for the paper Latent Visual Reasoning

SophiaVL-R1: Reinforcing MLLMs Reasoning with Thinking Reward

[NeurIPS 2024] Visual Perception by Large Language Model’s Weights

Tongyi Deep Research, the Leading Open-source Deep Research Agent

Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

A toolkit for developing and comparing reinforcement learning algorithms.

Muon is Scalable for LLM Training

SGLang is a fast serving framework for large language models and vision language models.

An Easy-to-use, Scalable and High-performance RLHF Framework based on Ray (PPO & GRPO & REINFORCE++ & TIS & vLLM & Ray & Dynamic Sampling & Async Agentic RL)

Open-source unified multimodal model

Enforce the output format (JSON Schema, Regex etc) of a language model

This is the della guide for Zhuang's group at Princeton University.

[ICCV 2025] Video-T1: Test-Time Scaling for Video Generation

A simple pip-installable Python tool to generate your own HTML citation world map from your Google Scholar ID.

EB1A DIY Collection

DIY for NIW/EB1A