Kyutai releases CADRE, 7 open causal diffusion video autoencoders for interactive world models
Kyutai released CADRE, a family of 7 open causal diffusion video autoencoders built for interactive world models. World models built on CADRE run at 20 fps on 1 H100 and react in under 230 ms, according to MarkTechPost's report.
Kyutai has released CADRE, a family of 7 open causal diffusion video autoencoders designed for interactive world models. World models built on CADRE run at 20 fps on 1 H100 GPU and react in under 230 ms, according to a report by MarkTechPost published October 23, 2026.
Key facts
- Developer: Kyutai
- Release: CADRE, a family of 7 open causal diffusion video autoencoders
- Throughput: 20 fps on 1 H100
- Reaction time: under 230 ms
- Use case: interactive world models
Parameter counts, benchmark scores, licensing terms and training data details were not part of the report's summary. Pricing is not applicable to the open release as described.
What this means
The two figures reported, 20 fps on a single H100 and a reaction time under 230 ms, are the performance numbers that matter for interactive use, where a world model has to respond to user input as it happens. Running on 1 H100 puts the reported setup within reach of a single high-end GPU. Because the 7 autoencoders are open, other developers can build world models on them.
Related Articles
MemTensor open-sources MemT 1.0 Flash and Nano, models with built-in trainable memory state
MemTensor has open-sourced MemT 1.0 Flash and Nano, models with a built-in trainable memory state that keeps long histories in a fixed state. The company reports about 4x faster inference at 256K history and 130x smaller storage.
Block announces Dessa, a speech-to-speech model trained from scratch on human conversation
Block announced Dessa, a speech-to-speech AI model it says was trained "from scratch on nothing but human conversation." The team says it can run locally on a MacBook Pro M4 Max "or better," and early access will go to a small group of AI safety researchers.
Kimi K3.1 Pro scores 55 on Artificial Analysis Index, fourth place behind Claude Opus 5.5, with API prices cut by a thir
Moonshot AI has released Kimi K3.1 Pro, which scores 55 on the Artificial Analysis Intelligence Index, an 11-point gain over Kimi K3 and fourth place overall. Parameter count is unchanged at 2.8 trillion, while API prices fall by 33 percent. Moonshot plans to release the weights by November 4, 2026.
MOSS-Image, a 5B open-weights image model, passes one million downloads in four days
MOSS-Image, an open-weights text-to-image model from MOSS (模思智能) and the OpenMOSS team, is built on a 5B-parameter DiT backbone and has passed one million downloads in four days, according to Chinese outlet QbitAI. It supports structured JSON layout input and text prompts up to 4096 tokens.
Comments
Loading...