gguf quantized version of cosmos-predict2 (incl. 2b and 14b)

  • drag cosmos to > ./ComfyUI/models/diffusion_models
  • drag t5xxl-old to > ./ComfyUI/models/text_encoders
  • drag pig to > ./ComfyUI/models/vae

screenshot

Prompt
On a rainy night, a girl holds an umbrella and looks at the camera. The rain keeps falling.
Prompt
On a rainy night, a girl holds an umbrella and looks at the camera. The rain keeps falling. a serene atmosphere
Prompt
On a rainy night, a girl holds an umbrella and looks at the camera. The rain keeps falling.
Prompt
On a rainy night, a girl holds an umbrella and looks at the camera. The rain keeps falling.
Prompt
The night of the city, from the perspective of the car's front window, real shooting, real scene, late-night emo style, emotional, sad, neon lights, sense of the future, rain, cinematic feel, cyberpunk
Prompt
The night of the city, from the perspective of the car's front window, real shooting, real scene, late-night emo style, emotional, sad, neon lights, sense of the future, rain, cinematic feel, cyberpunk

screenshot

  • don't need safetensors anymore; all gguf (model + encoder + vae)
  • full set gguf works on gguf-node (see the last item from reference below)
  • get more t5xxl-old gguf encoder either here or here
  • note (internal use) - ex: extra state excluded; ex3: ex + pos_embedder x 3 excluded

reference

Downloads last month
451
GGUF
Model size
14.3B params
Architecture
pig
Hardware compatibility
Log In to view the estimation

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Examples
Examples
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for calcuis/cosmos-predict2-gguf

Quantized
(1)
this model