Skip to main content
← Back to news

NVIDIA releases Cosmos3-Edge-Policy-DROID: a 4B policy for on-device robotics

Model releases
Source:NVIDIA (Hugging Face)
The 4B checkpoint of the Cosmos3 family turns natural-language instructions and visual observations from the DROID robot into action trajectories. NVIDIA's edge checkpoint base: OpenMDW 1.1 license, commercial and non-commercial use, released July 2026.
Read original article

Related articles

Cosmos 3 Edge: la policy DROID per la robotica che gira sull'edge

Cosmos3-Edge-Policy-DROID: la policy robotica da 4B di NVIDIA. Quanto hardware serve per farla girare in tempo reale sul Jetson, e dove si ferma.

Get the next local-AI briefing

Once a week, the top selection from iaLocale: articles, news and resources on local AI, and more.

After signing up you receive a confirmation email (double opt-in). Without confirmation we send nothing. No spam, we promise.

Related resources

Tools and resources from the iaLocale archive relevant to this content.

🤖

NVIDIA Cosmos3-Edge-Policy-DROID

Models

Post-trained 4B robot-policy checkpoint on the Cosmos3-Edge base: from natural-language instructions and visual observations of the DROID robot it generates action trajectories for manipulation tasks. Mixture-of-Transformers architecture (autoregressive tower + diffusion transformer), OpenMDW 1.1 license, commercial and non-commercial use.

🚗

NVIDIA Cosmos3-Nano-Sim-Auto

Models

Causal-distilled variant of the Cosmos3 family for real-time driving simulation: generates synchronized multi-camera video (up to 11 standard and fisheye views) and LiDAR, conditioned on a text prompt and WSM controls. 16B parameters, Mixture-of-Transformers architecture, OpenMDW 1.1 license, ready for commercial and non-commercial use.

📄

NVIDIA Nemotron Parse (famiglia OCR)

Models

NVIDIA's vision encoder-decoder family for local document understanding: markdown, LaTeX formulas, bounding boxes and a semantic class per element, in reading order. ViT-H (C-RADIO) + 10-block mBART decoder, all under 1 billion parameters. Locally via vLLM (v0.20–v0.26 for 2.0) or Transformers, on NVIDIA GPUs (Turing through Blackwell).

🖥️

NVIDIA A100 80GB (SXM)

Hardware

80GB HBM2e datacenter GPU: the standard for datacenter training and inference. Requires dedicated servers and enterprise power.

Browse all resources