Runway open-weights Praxis-1 27B world action model for robots, with NVIDIA
tlakomy · x · 2026-11-12
Runway released the open-weight Praxis-1 27B world action model, with NVIDIA highlighting deployment support across its GPUs — from data centers down to on-robot Jetson inference.
Resources include NVFP4 checkpoints on Hugging Face (runwayml/praxis-1-yam-nvfp4, Apache-2.0), a Jetson AI Lab tutorial, and post-training via NVIDIA NeMo Automodel. The NVFP4 variant applies quantization-aware training to the routed MoE expert layers, shrinking memory from 54.4GB to 17.5GB (experimental on SM120 GPUs like RTX 5090 / RTX PRO 6000 Blackwell). The model targets a bimanual YAM robot policy.
Related event: Runway Open-Sources Praxis-1 World Action Model, Tops RoboLab-120(14 posts)→
More from Embodied
- Apple affirms Vision Pro commitment; glasses rumored to be in secret development — Scobleizer · 2026-11-12
- ContactBench 1.0: A Benchmark for Contact-Rich Robot Manipulation With Printable Assets — leto__jean · 2026-11-12
- AMD runs MolmoAct2 robot policy locally on a Ryzen AI Max+ 395 mini PC — DJiafei · 2026-11-12
- Open-Source CoRL Paper Browser: 2,100+ Robot Learning Papers, a Decade of Searchable Research — promptdom · 2026-11-12
- World Labs and AMD enter robotics, Li Fei-Fei appears at CoRL — breadli428 · 2026-11-12
- SimEX: agents learn manipulation skills in imperfect simulation and transfer them to real robots — chris_j_paxton · 2026-11-12