{"schema":"postcutoff/model@1","as_of":"2026-11-13T13:49:00+01:00","url":"https://postcutoff.com/m/cosmos-3/","md":"https://postcutoff.com/m/cosmos-3/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"cosmos-3","name":"Cosmos 3 (Nano / Super)","org":"NVIDIA","family":"Cosmos","released":"2026-06-01","status":"current","type":"world-model","modality_in":["text","image","video","audio","action"],"modality_out":["text","image","video","audio","action"],"open_weights":true,"model_license":"OpenMDW-1.1 (commercial and non-commercial use)","context_window":null,"max_output":null,"knowledge_cutoff":null,"pricing":null,"price_line":"Price not published","access":[{"provider":"Hugging Face","model_id":"nvidia/Cosmos3-Nano","url":"https://huggingface.co/nvidia/Cosmos3-Nano"},{"provider":"Hugging Face","model_id":"nvidia/Cosmos3-Super","url":"https://huggingface.co/nvidia/Cosmos3-Super"},{"provider":"GitHub","url":"https://github.com/nvidia-cosmos","docs":"https://research.nvidia.com/labs/cosmos-lab/cosmos3/"}],"capabilities":[{"name":"Unified omni world model (generation + reasoning + action)","detail":"One Mixture-of-Transformers model (autoregressive + diffusion) replaces separate Cosmos Predict, Transfer, Reason and Policy models: world generation, physical reasoning, forward/inverse dynamics and action/policy generation.","first":true,"discovered":"launch","source":"https://nvidianews.nvidia.com/news/nvidia-and-global-robotics-leaders-take-physical-ai-to-the-real-world"},{"name":"Open omnimodal I/O","detail":"Inputs text, images, short video, audio and action trajectories (16-400 frames); outputs text, images, video (5-400 frames), 48 kHz stereo audio and actions (JSON).","first":false,"discovered":"launch","source":"https://huggingface.co/nvidia/Cosmos3-Nano"},{"name":"Leaderboard results","detail":"NVIDIA cites best open text-to-image and image-to-video models on Artificial Analysis and best policy model on RoboArena.","first":false,"discovered":"later","source":"https://www.nvidia.com/en-us/ai/cosmos/"}],"entry":"2026-06-01-nvidia-cosmos-3-open-release","notes":"Announced at GTC 2026-03-16 ('the first world foundation model unifying synthetic world generation, vision reasoning and action simulation' - NVIDIA claim); weights published 2026-05-31/06-01 (HF blog 'The First Open Omni-model for Physical AI Reasoning and Action'). Sizes: Nano 16B, Super 64B. Linux + Ampere/Hopper/Blackwell GPUs, BF16. Technical report dated 2026-06-22.","verified":"2026-09-29","body_md":"Sources: [HF blog](https://huggingface.co/blog/nvidia/cosmos-3-for-physical-ai), [Cosmos3-Nano](https://huggingface.co/nvidia/Cosmos3-Nano), [Cosmos3-Super](https://huggingface.co/nvidia/Cosmos3-Super), [technical report](https://research.nvidia.com/labs/cosmos-lab/cosmos3/technical-report.pdf).","page_url":"https://postcutoff.com/m/cosmos-3/","standing":{"kind":"world","newer_count":2,"newer":[{"id":"genie-4","name":"Genie 4","org":"Google DeepMind","released":"2026-10-27","page_url":"https://postcutoff.com/m/genie-4/"},{"id":"xiaomi-robotics-u0","name":"Xiaomi-Robotics-U0 (38B) / U0-4B","org":"Xiaomi","released":"2026-07-13","page_url":"https://postcutoff.com/m/xiaomi-robotics-u0/"}],"before":{"id":"cosmos-predict-2-5","name":"Cosmos Predict 2.5 / Transfer 2.5","org":"NVIDIA","released":"2025-10-06","page_url":"https://postcutoff.com/m/cosmos-predict-2-5/"},"after":null,"weights":"public","licence":"OpenMDW-1.1 (commercial and non-commercial use)","licence_permissive":false,"summary":"Weights are public. Licence: OpenMDW-1.1. See the licence for commercial terms."}}