French AI research lab Kyutai has introduced CADRE, a family of open causal diffusion representation autoencoders built for interactive world models. The release is aimed at reducing action latency, a key constraint for systems that need to react quickly to user input or changing environments. According to Kyutai, its world model runs at 20 fps on a single NVIDIA H100 GPU and delivers response times below 230 milliseconds.
The model family includes seven autoencoders, each with about 476 million trainable parameters, and uses a frozen 86 million-parameter DINOv3 ViT-B/16 model. Kyutai said the full world model totals 1.2 billion parameters. In benchmark comparisons against MIRA, Wan 2.2, Cosmos, and LTX-2, CADRE outperformed rivals in 42 of 48 tests. In a driving-scene FVD result, CADRE-2×16×16 with single-step decoding posted 162.6, compared with 599.7 for Wan 2.2. The system was trained on a mixed dataset spanning 23 public video sources and roughly 76,000 hours of data, including web video, robotics, driving, and gameplay footage.
French AI research lab Kyutai has released CADRE, a family of open causal diffusion representation autoencoders designed for interactive world models. The project targets action latency, which Kyutai identified as a core bottleneck for this type of system.
Kyutai said its world model can run at 20 fps on a single H100 GPU, with response time kept below 230 milliseconds.
Model size and architecture
CADRE includes seven autoencoders, each with about 476 million trainable parameters. It also uses a frozen DINOv3 ViT-B/16 model with 86 million parameters. The full world model has 1.2 billion parameters in total.
Benchmark results
In 48 comparisons against MIRA, Wan 2.2, Cosmos, and LTX-2, CADRE delivered better results in 42 cases.
For the FVD metric in driving scenarios, CADRE-2×16×16 with single-step decoding reached 162.6, versus 599.7 for Wan 2.2.
Training data
The model was trained on a mixed dataset built from 23 public video sources, covering about 76,000 hours in total. The data includes web video, robotics, driving, and game footage.
The item was published by Techub and cited MarkTechPost in the source summary.
This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan. Disclaimer:
The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.
Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.
