Apple releases FastFrames open video models that generate 480p video at interactive speed
Apple has released FastFrames, two open video generation models that produce 480p video at interactive speed, on Hugging Face alongside a research paper.
FastFrames-S has 1.3 billion parameters and FastFrames-L has 14 billion. Both are autoregressive, two-step video diffusion models. Apple's published timings put one generation step for FastFrames-S at about 35 milliseconds on a single GPU at 480p, in line with the 36 milliseconds reported in the paper.
Apple also released an optimised version of the LingBot 1.3B model that it says is 16 times faster than the official version and allows fluid, camera-controlled generation, with an interactive demo. The release includes code for latency profiling, text-to-video generation, VBench evaluation and training, and runs on Linux with an Nvidia GPU.
The weights for both models are available on Hugging Face, with the project page and arXiv paper linked from the model card.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Daily AI Brief.
Also, consider following us on social media:
Daily AI Brief
Daily report covering major AI developments and industry news, with both top stories and complete market updates
Industry analysis
2025 Global Business Services Agenda: Gen AI Takes Center Stage
This industry analysis by The Hackett Group explores the transformative impact of generative artificial intelligence (Gen AI) on global business services (GBS) in 2025. The study highlights the shift from exploration to acceleration of Gen AI initiatives, with 89% of executives advancing these projects to improve customer satisfaction, innovate products, and reduce costs. The report also discusses the challenges and strategies for successful Gen AI adoption, emphasizing the need for a technology-enabled operating model and the importance of reskilling the workforce.
Read moreYou may also like
Fal.ai Releases FLUX 3 Max Video Model
Liquid AI Releases LFM3, Six Open-Weight Models for On-Device AI
Tencent Releases Hy-Nomas for Interactive Video World Generation
Liquid AI releases LiquidCodec, an open 74M-parameter streaming speech codec
StepFun releases Step 5 Preview open weights
Daily AI Brief: the AI news that matters, in your inbox.