USASI
Model release

NVIDIA Cosmos3-Nano

Release in the NVIDIA Cosmos family · version Cosmos3-Nano (16B)

Maintained by NVIDIA1

Cosmos3-Nano is the 16B-parameter model in NVIDIA's Cosmos 3 family, released on May 31, 2026. It shares the family's mixture-of-transformers design, generating text autoregressively and images, video, audio, and robot actions by diffusion, from combinations of text, image, video, audio, and action-trajectory inputs.19

Last reviewedEntry updated Documented release May 31, 2026

Availability and license

Overall availability

Public

Weights are downloadable from Hugging Face without an access gate under OpenMDW-1.1; the card states the model is ready for commercial and non-commercial use and lists global deployment geography.21

Availability is separate from permission: read the license before using or redistributing.

OpenMDW-1.1 grants permission to deal in the model, its parameters, and related data, documentation, and software without restriction under copyright, patent, database, and trade secret rights. Distributors must keep a copy of the agreement and applicable notices, and rights end for anyone who sues claiming the materials infringe a patent or copyright, unless responding to a suit brought first. It places no restrictions on outputs. It is not on this catalog's OSI-approved list.34

Model-disclosure tier

Computed from the checklist below using USASI rubric v0.1. An editorial category, not a certification.
Model-disclosure tier (USASI rubric v0.1): Open-weight

The model parameters for this release can be downloaded by the public. License terms may still restrict use, redistribution, or commercial use.

The weights are under a license that is not on the rubric's OSI-approved list. Read its terms before use.

How tiers are computed

Public materials checklist

Items for a model under USASI rubric v0.1. Unknown means unassessed or insufficient evidence.
Public materials checklist for NVIDIA Cosmos3-Nano
ItemStatusNotes and evidence
WeightsCan the general public download the model parameters for this release?PublicWeights are published on Hugging Face without an access gate.21
Inference codeIs code for running the model published?PublicThe card documents inference with vLLM-Omni, vLLM (through a Cosmos-Framework plugin package), Hugging Face Diffusers, and SGLang Diffusion.16
Training codeIs the code used to train the model published?PartialCosmos-Framework publishes a distributed trainer whose documented recipes cover supervised fine-tuning (including a Nano vision SFT recipe). The report describes the internal training infrastructure used for both towers but does not identify it as Cosmos-Framework, and pre-training configurations were not found.679
Training-data informationPublic = the training data itself can be obtained. Partial = composition or sources are documented without full access.PartialThe card summarizes the corpus (about 1.3B data points from 393 dataset entries, drawn from NVIDIA-owned data and public datasets, plus synthetic data) and links a public summary of training content. The report lists five synthetic datasets released on Hugging Face; most training data is not released.19
Training recipeAre the training configuration and procedure documented in enough detail to follow?PublicThe technical report describes the training stages, data mixtures, optimizer settings, learning-rate schedules, and the token and GPU counts used for Nano.9
Evaluation materialsPublic = evaluation code or prompts that let others re-run the evaluations are published. Partial = results only.PartialThe card and report publish benchmark results, and the report releases one human evaluation benchmark (Cosmos-HUE); code to re-run all reported evaluations was not verified.19

What it is useful for

The card lists physical AI uses in robotics, autonomous vehicles, and industrial and smart spaces; NVIDIA positions Nano for fast video and action reasoning and as a balance of speed and quality across deployments.1108

Run and use notes

Documented facts only. No hardware or performance claims are made without a cited source and stated assumptions.
  • The card lists PyTorch, vLLM-Omni, Diffusers, and SGLang as runtimes; NVIDIA Ampere, Hopper, and Blackwell GPUs; and Linux as the tested operating system. It states that only BF16 precision is tested and gives a recommended vLLM-Omni serving command for an H200 GPU.1

Organization context

Provenance and derivatives

Trained by NVIDIA. The technical report states that Cosmos3-Nano adapts the Qwen3-VL 8B architecture and is initialized from pre-trained Qwen3-VL-8B weights (developed by the Qwen team at Alibaba Cloud), and that visual generation uses the video VAE encoder from Wan2.2-TI2V-5B (published on Hugging Face by Wan-AI). The card lists synthetic training data generated with HiDream-I1, Qwen-Image-2512, and Qwen3-VL.911211

Other releases in the NVIDIA Cosmos family

NVIDIA Cosmos family overview

U.S. eligibility

Project eligibility rests on documented governing or maintaining entities, not on contributors.

Eligible · basis: U.S. headquarters

The model card names NVIDIA as the model developer. NVIDIA's principal executive offices are in Santa Clara, California, per its Form 10-Q for the quarter ended July 26, 2026. The model is initialized from third-party Qwen3-VL weights, recorded under provenance; that base is not U.S.-developed.113912

Assessed Sep 29, 2026

Sources

  1. 1.
    nvidia/Cosmos3-Nano model card (external site: huggingface.co)

    NVIDIA (Hugging Face) · Model card · accessed Sep 29, 2026

  2. 2.
  3. 3.
    OpenMDW License 1.1 (external site: openmdw.ai)

    OpenMDW · License · accessed Sep 29, 2026

  4. 4.
  5. 5.
    NVIDIA/cosmos-framework LICENSE (external site: raw.githubusercontent.com)

    NVIDIA (GitHub) · License · accessed Sep 29, 2026

  6. 6.
    NVIDIA/cosmos-framework README (external site: raw.githubusercontent.com)

    NVIDIA (GitHub) · Repository · accessed Sep 29, 2026

  7. 7.
  8. 8.
    NVIDIA/Cosmos (GitHub) (external site: github.com)

    NVIDIA (GitHub) · Repository · accessed Sep 29, 2026

  9. 9.
  10. 10.
    NVIDIA Launches Cosmos 3, the Open Frontier Foundation Model for Physical AI (external site: nvidianews.nvidia.com)

    NVIDIA Newsroom · Announcement · published May 31, 2026 · accessed Sep 29, 2026

  11. 11.
    Wan-AI/Wan2.2-TI2V-5B model card (external site: huggingface.co)

    Wan-AI (Hugging Face) · Model card · accessed Sep 29, 2026

  12. 12.
    QwenLM/Qwen3-VL (GitHub) (external site: github.com)

    Qwen team (GitHub) · Repository · accessed Sep 29, 2026

  13. 13.
    NVIDIA Corporation Form 10-Q for the quarter ended July 26, 2026 (external site: sec.gov)

    U.S. Securities and Exchange Commission (EDGAR) · Filing · accessed Sep 29, 2026

This listing is not an endorsement, a safety assessment, or a federal approval.

Support Us

Help keep USASI useful.

Optional. No USASI account required. Payment takes place on the linked provider’s website (Buy Me a Coffee).

About supporting this project