Thrilled to share recent research from our fascinating lab members and collaborators at #ICLR2025! 🚀✨ Come say hi in our poster sessions and dive into discussions on LLM agents, reasoning, long-context training, efficient inference, and more. We’re excited to share, learn and

STA is accepted by ICML 2025!!

Announcing FastVideo V1, a unified framework for accelerating video generation. FastVideo V1 offers: - A simple, consistent Python API - State of the art model performance optimizations - Optimized implementations of popular models Blog: hao-ai-lab.github.io/blogs/fastvide…

amazing！

I will be giving a talk in GPU MODE tomorrow (May 31 12pm PST) about FastVideo/STA/VSA. Come if you're interested! youtube.com/watch?v=x44iGp…

🔧🤖 New wave of open-source LLMs like Deekseek-R1-0528 and Qwen3-235B-A22B are leveling up with stronger agentic performance. We test them in head-to-head gameplay — the upgraded Deekseek-R1-0528 outsmarts strong reasoning models like o4-mini across several games and it nearly

🚀 Attention is the bottleneck in video DiTs—5 s of 720p = 100K+ tokens, quadratic cost blows up fast. Sparse/linear attention is 🔑 for long-context world models. 🧠 Track relavent papers in our awsome-video-attention repo → github.com/hao-ai-lab/Aws… #WorldModel #VideoAI

Heading to ICML next week (Monday - Thursday). Down to chat research, ideas, anything cool, or just hang 😄📍🎯

📣 We’ve had three papers accepted at #ICML2025, Hao-AI-Lab is sending Hao Zhang to attend ICML in person😂! If you're around, please find Hao at the venue and chat with him about video diffusion, LLM agents, and efficient attention 👋🧠 🎬 Fast Video Generation with Sliding

Watch my talk about NATTEN on GPU MODE this Saturday at 3PM ET / noon PT. I'll go over all the exciting new features we shipped very recently, especially our Hopper and Blackwell FNA kernels, now speeding up video / world models by up to 2.6X e2e! youtube.com/watch?v=mF_H_J

I learned a lot from NATTEN!

(1/n) 🚀 With FastVideo, you can now generate a 5-second video in 5 seconds on a single H200 GPU! Introducing FastWan series, a family of fast video generation models trained via a new recipe we term as “sparse distillation”, to speed up video denoising time by 70X! 🖥️ Live

Crazy fast!! Great work from Hao AI Lab

Try FastWan at fastwan.fastvideo.org!

Simple design ofen wins in the long run. GPT-OSS uses sliding window atteniton. Our Sliding Tile Attention brings efficieint window attention to video generation: arxiv.org/abs/2502.04507

[Lmgame Bench] 🔥 We tested Openai’s GPT-5-thinking-high and two recent open-source models in our Lmgame Bench! Across 26 models and 6 games (Sokoban, Tetris, 2048, Candy Crush, Mario, Ace Attorney), Here’s where they landed: GPT-5-thinking-high → #2

[Lmgame Bench] 🤔 Ever wondered how to evaluate different games in Lmgame-Bench or even add your own, but don’t know where to start? We’ve made it super easy to run evaluations and integrate new games. Our latest blog walks you through a few key features from Lmgame Bench

Excited to share my 1st project as a Research Scientist Intern at Meta FAIR! Grateful to my mentor Jiawei Zhao for guidance, and to Yuandong Tian & Xuewei for their valuable advice and collaboration. Our work DeepConf explores local confidence for more accurate & efficient LLM reasoning!

[1/5] [Lmgame Bench] 🎮 Question: Can RL-based LLM post-training on games generalize to other tasks? We shared a preliminary study to explore this question: - Same-family (in-domain): Training on 6×6 Sokoban → 8×8 and Tetris (1 block type) → Tetris (2 block types) transfers,

🚀 Thrilled to share that our lab has THREE papers accepted at #NeurIPS2025 on AI efficiency from reasoning to video generation. Come hang out with us, it's going to be a lot of fun this year here local to UCSD! 😎 📊 Efficiently Scaling LLM Reasoning with Certaindex Introduces