FastH3 by FastVideo is now available in ComfyUI Video and native stereo audio, generated together, in seconds. Best for: → Previz and animatics that need lots of takes, quickly → Timing and dialogue tests before committing to a full-quality render → Short-form and social work on tight turnarounds Run it locally in ComfyUI and soon on Cloud ⬇️

I gave Astra one brief and it delivered the whole scene in Blender, camera direction included. All of it ran inside MiniMax Design. Astra sits there as the agent and talks to Blender over an official connector, so everything syncs straight onto the canvas. The final video is MiniMax H3, from the same canvas. #MiniMaxdesign - https://design.minimax.io/

I used the work generated with GPT-6 Astra to build out a set, a story. Now I'm bringing it to life with @Hailuo_AI Minimax H3. Here's a tip that's worked well for me 👇

SGLang-Diffusion with VDN-H3 now generates 14.4s of 768p video in just 9.0s 🚀 On 8× B200, 8 step denoising takes just 6.9s, reaching over 2× real time. The 9.0s figure covers the full generation request after warmup. No measured quality regression versus dense 50-step H3 across 103 test prompts. 🧵

We Must Pace the Frontier

Final call. Bestiary closes tonight, Sep 14 at 23:59 (PT). ⏳ $8,000 cash, 200,000 Credits, and ten Audience Choice awards are still on the table — and they go to the people who actually hit submit. One strange, beautiful short film, 30 seconds or longer, generated with MiniMax H3. Any myth, any era, any world you can dream up. The bestiary doesn't close itself. Finish your film before the clock runs out.

GPT-6 Astra can now generate animatable Gaussian Splat characters. We connected it to the PINOC MCP and asked for a backrooms-style, Exit 8-ish game. We described the character we wanted and the motions. [MCP link in the comment 👇] Astra generated the character and every motion through PINOC through free preset animations and text to animation, and wrote the loop and the anomaly logic itself, and shipped the whole thing in a few sessions.

🐶🏁 Three models. One race. We independently tested H3 Max by @fal, minimax H3 by @MiniMax_AI, and FastH3 preview by @haoailab @haozhangml @wlsaidhi across robotics, animation, movies, and ads. H3 Max takes the overall lead 🏆. FastH3_preview, meanwhile, is a community OSS effort that’s already keeping up surprisingly well, even beating H3 on overall Prompt Adherence 🥳 The bigger gaps show up in Visual Integrity and Human Preference. Full evaluation: https://physionlabs.ai/blog/minimax-h3-evaluation

🎬Video generation faster than playback! 🚀MiniMax H3 on vLLM-Omni + FastVideo's FastH3: a complete 10.1s MP4 - video AND synchronized audio - rendered in 8.7s!⚡️ Thanks to @MiniMax_AI for the great Minimax H3 release, the FastVideo team @haoailab for open-sourcing FastH3 and helping on the serving integration, and @NVIDIAAI for the continued sponsorship and joint optimization efforts!

Introducing Atlas: The world's first multimodal world model that generates image and video frames with pixel-perfect camera control and reconstructs them in 3D. Model the world, move the camera, and simulate space & time.

At Reactor we ❤️ open-source. We teamed up with @haoailab to ship an infinite live stream powered by FastVideo’s FastH3: https://twitch.tv/dereactor And we’re open-sourcing everything!

🚀 What if video generators could build on representations that already understand the visual world? We are excited to introduce V-RAE: Rethinking Video Latent Spaces for Generation. Recent progress in image generation has begun to move beyond conventional VAE latents, exploring both direct pixel-space and representation-based approaches. Video generation, however, still depends heavily on latent compression, as the scale and redundancy of spatiotemporal data make direct modeling prohibitively expensive. However, most video VAEs are optimized for pixel reconstruction, and a latent space that reconstructs well is not necessarily easy to generate. V-RAE takes a different approach: it directly uses representations from frozen vision foundation models as the generative latent space, rather than as auxiliary supervision. We study DINOv3, SigLIP2, EUPE, and V-JEPA 2.1. A lightweight temporal attention pooling module compresses their dense features by 4×, followed by a spatiotemporal Transformer decoder. Under matched generation backbones, latent budgets, and training settings, V-RAE achieves: 🏆 2.13 rFVD on Kinetics-600 🎬 117.86 gFVD on UCF101 and 19.16 gFVD on Kinetics-600 ⚡ Up to 6× faster convergence than VAE-based latent spaces 🧠 90.92% semantic probing accuracy on UCF101 🌍 Better future prediction on Cityscapes, reducing gFVD from 144.47 to 111.36 Our experiments also reveal a broader finding: Good Reconstruction ≠ Good Generation. During generation, predicted latents inevitably deviate from real encoding trajectories. If the latent space is not sufficiently smooth, small errors can be amplified into visible artifacts. We therefore introduce tFVD to evaluate temporal smoothness and robustness to latent prediction errors. It correlates much more strongly with downstream generation quality, reaching 0.919 on Kinetics-600. The takeaway: A latent space is not merely where videos are compressed—it determines what the generator must learn. When semantics and temporal structure are already organized in the representation, generation becomes easier to learn. Representation first. Generation follows. Many thanks to my mentors, @ScottNLP and @SQWu_Tori, for their continuous guidance and support. I am also deeply grateful to @sainingxie for his valuable guidance and invaluable feedback, which greatly helped shape V-RAE. 🙏 Hi @_akhaliq, we would truly appreciate your help in sharing V-RAE with the broader AI research community. Thank you! 🙏 📄 Paper: https://arxiv.org/abs/2608.13556 💻 Code: https://github.com/V-RAE/V-RAE 🤗 Models: https://huggingface.co/Guomh0707/V-RAE-Models 🌐 Project: https://v-rae.github.io #VideoGeneration #GenerativeAI #ComputerVision #WorldModels #RepresentationLearning #RAE

Can MLLMs actually track what's happening in a video? Introducing VSTAT 🎯, our new benchmark for visual state tracking. The tasks are simple: count cups, read typed words, count page flips. Humans solve them easily. MLLMs don't. https://vision-x-nyu.github.io/vstat-site/ 🧵 [1/11]