Cosmos is NVIDIA's language model. NVIDIA's world foundation model for physical AI, designed to generate and simulate realistic physical world states for robotics and autonomous systems.
Specifications
Canonical IDnvidia-cosmos
TypeLanguage
StatusActive
CreatorNVIDIANVIDIA
Input ModalitiesText
Output ModalitiesText

Capabilities

Input1/5
Text
Image·
Audio·
Video·
PDF·
Output1/5
Text
Image·
Audio·
Video·
Embedding·
Capabilities0/13
Reasoning·
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs·
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·

Versions

VersionReleasedContextInput / 1MOutput / 1MStatus
CosmosCurrent
Cosmos Reason1 7BAvailable

Model IDs