Hugging Face's logo

nvidia
/
Cosmos3-Super-Image2Video-4Step

Image-to-Video
Cosmos
Diffusers
Safetensors
cosmos3_omni
nvidia
cosmos3
vllm-omni
video-generation
Model card Files
xet
 Community
8
tangyue0820 commited on
Commit
4848ed6
·
verified ·
1 Parent(s): 237df5c

Add Cosmos3 family tree graph

Browse files

Adds the Cosmos3 product and examples lineage graph to the model overview, following the format used by the recent Cosmos3 Nano simulation model cards. Please review the graph and lineage description.

Files changed (3) hide show
  1. .gitattributes +1 -0
  2. README.md +4 -0
  3. images/cosmos3-super-image2video-4step-graph.png +3 -0
.gitattributes CHANGED
@@ -39,3 +39,4 @@ assets/example_output.mp4 filter=lfs diff=lfs merge=lfs -text
39
  images/aa-i2v-all-models-2026-07-17.png filter=lfs diff=lfs merge=lfs -text
40
  images/aa-i2v-open-models-2026-07-17.png filter=lfs diff=lfs merge=lfs -text
41
  assets/example_output_diffusers.mp4 filter=lfs diff=lfs merge=lfs -text
 
 
39
  images/aa-i2v-all-models-2026-07-17.png filter=lfs diff=lfs merge=lfs -text
40
  images/aa-i2v-open-models-2026-07-17.png filter=lfs diff=lfs merge=lfs -text
41
  assets/example_output_diffusers.mp4 filter=lfs diff=lfs merge=lfs -text
42
+ images/cosmos3-super-image2video-4step-graph.png filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -20,6 +20,10 @@ tags:
20
 
21
  # Model Overview: Cosmos3-Super-Image2Video-4Step
22
 
 
 
 
 
23
  ## Description
24
 
25
  Cosmos3-Super-Image2Video-4Step is a 4-step distilled version of the base [Cosmos3-Super-Image2Video](https://huggingface.co/nvidia/Cosmos3-Super-Image2Video) model. Given an input image plus an optional text prompt, it generates a temporally coherent video.
 
20
 
21
  # Model Overview: Cosmos3-Super-Image2Video-4Step
22
 
23
+ ![Cosmos3 product and examples lineage highlighting Cosmos3-Super-Image2Video-4Step](images/cosmos3-super-image2video-4step-graph.png)
24
+
25
+ The highlighted checkpoint, Cosmos3-Super-Image2Video-4Step, is a step-distilled variant of Cosmos3-Super-Image2Video.
26
+
27
  ## Description
28
 
29
  Cosmos3-Super-Image2Video-4Step is a 4-step distilled version of the base [Cosmos3-Super-Image2Video](https://huggingface.co/nvidia/Cosmos3-Super-Image2Video) model. Given an input image plus an optional text prompt, it generates a temporally coherent video.
images/cosmos3-super-image2video-4step-graph.png ADDED

Git LFS Details

  • SHA256: 3bb6cf8ebdf49c74011b64e17777a75d38718d3cf214c0fa31303c125685c5a2
  • Pointer size: 131 Bytes
  • Size of remote file: 149 kB