I have engine or simulator renders and want photoreal video through NVIDIA Cosmos-Transfer: what do I feed it, what comes out, and what do the control weights do?
Cosmos-Transfer1-7B (paper, Mar-Apr 2025) and Cosmos-Transfer2.5-2B (Nov 2025 report, repository docs) take a text prompt, an input video and one to four per-modality control videos (blur, Canny edge, depth, segmentation; LiDAR and HD-map in AV variants), each with a control_weight in [0, 1] and optionally a spatiotemporal mask. The vendor guidance is that higher weights follow the control more strictly at some cost in quality, and that about 0.5 is a middle starting point. Missing depth/segmentation controls are computed from the input video, so engine-exported passes are an override path, not a requirement.
Cosmos-Transfer1-7B (paper, Mar-Apr 2025) and Cosmos-Transfer2.5-2B (Nov 2025 report, repository docs) take a text prompt, an input video and one to four per-modality control videos (blur, Canny edge, depth, segmentation; LiDAR and HD-map in AV variants), each with a control_weight in [0, 1] and optionally a spatiotemporal mask. The vendor guidance is that higher weights follow the control more strictly at some cost in quality, and that about 0.5 is a middle starting point. Missing depth/segmentation controls are computed from the input video, so engine-exported passes are an override path, not a requirement.
Curator qualifications: All excerpts located. Cosmos-Transfer inputs (prompt, input video, 1-4 control videos) documented; weights terms are NVIDIA's (see C3R Cosmos guardrail L00 decision). Rights/verification notes are shown as limitations (L00 0019/ADR-0011): they restrict reuse of the cited material or record what was not verified; they are not cleared by this summary.
Limitations: [component_reuse_restriction] NVIDIA Open Model License terms per earlier Cosmos reviews; no new rights reading.
Evidence status: statements located in the cited sources by a correlated desk review (Claude); not executed; not independently reviewed.
Structured fields
{
"citations": [
{
"url": "https://arxiv.org/abs/2503.14492v2",
"note": "candidate evidence excerpts"
},
{
"url": "https://github.com/nvidia-cosmos/cosmos-transfer1/blob/5005e823dbd478ad8e51f6bc28a913a13a994b5f/examples/inference_cosmos_transfer1_7b.md",
"note": "candidate evidence excerpts"
},
{
"url": "https://github.com/nvidia-cosmos/cosmos-transfer2.5/blob/2ff49d0717af02057ae79bc75c00fbff9da1b4e7/docs/inference.md",
"note": "candidate evidence excerpts"
},
{
"url": "https://github.com/nvidia-cosmos/cosmos-transfer1/blob/5005e823dbd478ad8e51f6bc28a913a13a994b5f/README.md",
"note": "candidate evidence excerpts"
}
]
}Evidence claims
- [sourced_fact] Cosmos-Transfer1-7B (paper, Mar-Apr 2025) and Cosmos-Transfer2.5-2B (Nov 2025 report, repository docs) take a text prompt, an input video and one to four per-modality control videos (blur, Canny edge, depth, segmentation; LiDAR and HD-map in AV variants), each with a control_weight in [0, 1] and optionally a spatiotemporal mask. The vendor guidance is that higher weights follow the control more strictly at some cost in quality, and that about 0.5 is a middle starting point. Missing depth/segmentation controls are computed from the input video, so engine-exported passes are an override path, not a requirement. · locator: candidate evidence excerpts source
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| documentation | unknown | permitted | permitted | unknown | unknown | Applies to this original summary only, not to the cited sources. |
Linked records
- cites → https://arxiv.org/abs/2503.14492v2
- cites → https://github.com/nvidia-cosmos/cosmos-transfer1/blob/5005e823dbd478ad8e51f6bc28a913a13a994b5f/README.md
- cites → https://github.com/nvidia-cosmos/cosmos-transfer1/blob/5005e823dbd478ad8e51f6bc28a913a13a994b5f/examples/inference_cosmos_transfer1_7b.md
- cites → https://github.com/nvidia-cosmos/cosmos-transfer2.5/blob/2ff49d0717af02057ae79bc75c00fbff9da1b4e7/docs/inference.md