Introducing H3 Max by fal

Announcing H3 Max, a post-trained version of MiniMax H3 developed by fal Research and optimized for maximum speed by fal's inference team. In our human preference evaluations, H3 Max ranks #1 across overall quality, prompt understanding, and aesthetics against leading video models. It does so while generating a

Announcing H3 Max, a post-trained version of MiniMax H3 developed by fal Research and optimized for maximum speed by fal's inference team. In our human preference evaluations, H3 Max ranks #1 across overall quality, prompt understanding, and aesthetics against leading video models. It does so while generating a 5-second video in under 3 seconds, which is roughly 35x the throughput of the official MiniMax H3 endpoint, and on average 15x faster than anything with comparable quality. These results challenge a common tradeoff in generative video: that higher quality has to come at the cost of…

saved by

• Varun Goyal

related reading

• MiniMax H3: Open Weights & Prompting Techniques Complete Guide.ghost.oxen.ai
• The First Fully General Computer Action Model | blogsi.inc
• Kimi K3 Tech Blog: Open Frontier Intelligencekimi.com
• H3: Language Modeling with State Space Models and (Almost) No Attention · Hazy Researchhazyresearch.stanford.edu
• Fable is SOTA at CIFAR Speedrun (& specification gaming): lessons on AI R&D automationfulcrum.inc
• Look Ma, No Bubbles! Designing a Low-Latency Megakernel for Llama-1B · Hazy Researchhazyresearch.stanford.edu
• Optimizing inference · Hugging Facehuggingface.co
• FLUX 3 x mimic: The Next Generation of Video-Action Modelsbfl.ai
• Kimi-K2.5 Inference Benchmark - Luminalluminal.com
• Veo 3.1deepmind.google
• PostTrainBenchposttrainbench.com
• Mosaic LLMs: GPT-3 quality formosaicml.com