Cambrian-S: Towards Spatial Supersensing in Video

Moving beyond the traditional paradigms of “Thinking with Text” (e.g., Chain-of-Thought) and “Thinking with Images”, we propose “Thinking with Video”. It naturally enables human-like dynamic reason…

Official implementation of "Continuous Autoregressive Language Models"

RLinf is a flexible and scalable open-source infrastructure designed for post-training foundation models (LLMs, VLMs, VLAs) via reinforcement learning.

The official repository for the paper "ThinkMorph: Emergent Properties in Multimodal Interleaved Chain-of-Thought Reasoning"

This is the official repo for the paper "LongCat-Flash-Omni Technical Report"

Native Multimodal Models are World Learners

Official repo of Toucan: Synthesizing 1.5M Tool-Agentic Data from Real-World MCP Environments

AgenTracer: A Lightweight Failure Attributor for Agentic Systems

"ViMax: Agentic Video Generation (Director, Screenwriter, Producer, and Video Generator All-in-One)"

On the Effect of Instruction Tuning Loss on Generalization

Official implementation of "Grasp Any Region: Towards Precise, Contextual Pixel Understanding for Multimodal LLMs".

Efficient Mixture of Experts for LLM Paper List

Source code of "Dr.LLM: Dynamic Layer Routing in LLMs"

Official Implement of "Agentic Jigsaw Interaction Learning for Enhancing Visual Perception and Reasoning in Vision-Language Models"

"LightAgent: Lightweight and Cost-Effective Mobile Agents"

Demystifying Reinforcement Learning in Agentic Reasoning

Official repo for "Spotlight on Token Perception for Multimodal Reinforcement Learning"

AgentFlow: In-the-Flow Agentic System Optimization

The Official Implementation of DIVER

Official codebase for "Self Forcing: Bridging Training and Inference in Autoregressive Video Diffusion" (NeurIPS 2025 Spotlight)

🔥🔥🔥 Latest Papers, Codes and Datasets on Video-LMM Post-Training

Thinking with Videos from Open-Source Priors. We reproduce chain-of-frames visual reasoning by fine-tuning open-source video models. Give it a star 🌟 if you find it useful.

Official Repo for Self-Forcing++ High Quality Long Video Generation

Qwen3-omni is a natively end-to-end, omni-modal LLM developed by the Qwen team at Alibaba Cloud, capable of understanding text, audio, images, and video, as well as generating speech in real time.