A generalist video MLLM built for fine-grained motion, long-form reasoning, temporal grounding, and online proactive response.

TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMs

Evaluation code for "Benchmarking Visual State Tracking in Multimodal Video Understanding"

Official implementation of EgoHOD at ICLR 2025; 14 EgoVis Challenge Winners in CVPR 2024

Official implementation of "HowToCaption: Prompting LLMs to Transform Video Annotations at Scale." ECCV 2024

[ICCV2023] EgoObjects: A Large-Scale Egocentric Dataset for Fine-Grained Object Understanding

[Nips 2025] EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Video Generation

CVPR '26 Highlight

TIPSv2 (CVPR'26) and TIPS (ICLR'25)

[CVPR 2026 Highlight] A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens

[ICML 2026] Official implementation of "Deep Forcing: Training-Free Long Video Generation with Deep Sink and Participative Compression"

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

Cosmos Policy

Code to pretrain, fine-tune, and evaluate DreamZero and run sim & real-world evals

Official Python inference and LoRA trainer package for the LTX-2 audio–video generative model.

Frontier Multimodal Foundation Models for Image and Video Understanding

Accepted By The 39th Annual Conference on Neural Information Processing Systems Datasets and Benchmarks Track

Official code for MotionBench (CVPR 2025)

[NeurIPS 2025 Oral]Infinity⭐️: Unified Spacetime AutoRegressive Modeling for Visual Generation

Improving Motion in Image-to-Video Models via Adaptive Low-Pass Guidance (CVPR 2026 Highlight)

[RSS 2023] Diffusion Policy Visuomotor Policy Learning via Action Diffusion

MotionStream: Real-Time Video Generation with Interactive Motion Controls

Learning Real-World Action-Video Dynamics with Heterogeneous Masked Autoregression

Official PyTorch implementation of "Video Summarization with Large Language Models" (CVPR 2025).

Native Multimodal Models are World Learners

[ICML2025] The code and data of Paper: Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generation

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

Video Generation, Physical Commonsense, Semantic Adherence, VideoCon-Physics

Long Video Gen Infrastructure