File size: 2,588 Bytes
b22f33a
 
 
 
 
 
 
 
 
 
14e2b08
 
 
 
 
 
 
 
 
 
 
 
 
 
734c2e0
 
 
 
 
 
 
 
c6b8080
 
 
 
b22f33a
5e860b6
b22f33a
 
 
9bafd43
 
7cdcb2e
14e2b08
 
29759d8
0ab7a3d
c038d33
 
34a6a22
0d8dd81
c038d33
7cdcb2e
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
---
license: other
license_name: nvidia-open-model-license
license_link: >-
  https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/
base_model:
- nvidia/Cosmos-Predict2-2B-Text2Image
pipeline_tag: text-to-image
tags:
- gguf-node
widget:
- text: >-
    On a rainy night, a girl holds an umbrella and looks at the camera. The rain keeps falling.
  output:
    url: samples\Cosmos__00001_.png
- text: >-
    On a rainy night, a girl holds an umbrella and looks at the camera. The rain keeps falling.
    a serene atmosphere
  output:
    url: samples\Cosmos__00002_.png
- text: >-
    On a rainy night, a girl holds an umbrella and looks at the camera. The rain keeps falling.
  output:
    url: samples\Cosmos__00003_.png
- text: >-
    On a rainy night, a girl holds an umbrella and looks at the camera. The rain keeps falling.
  output:
    url: samples\Cosmos__00004_.png
- text: >-
    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
  output:
    url: samples\Cosmos__00005_.mp4
- text: >-
    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
  output:
    url: samples\Cosmos__00006_.mp4
---
# **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](https://raw.githubusercontent.com/calcuis/comfy/master/cosmo-predict2-t2i.png)

<Gallery />

![screenshot](https://raw.githubusercontent.com/calcuis/comfy/master/cosmos-predict2-v2w.png)

- 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](https://huggingface.co/calcuis/pig-encoder/tree/main) or [here](https://huggingface.co/chatpig/t5-v1_1-xxl-old-encoder-gguf/tree/main)
- note (for internal use) - `ex`: extra state excluded; `ex3`: ex + pos_embedder x 3 excluded

### **reference**
- base model from [nvidia](https://huggingface.co/nvidia)
- comfyui from [comfyanonymous](https://github.com/comfyanonymous/ComfyUI)
- gguf-node ([pypi](https://pypi.org/project/gguf-node)|[repo](https://github.com/calcuis/gguf)|[pack](https://github.com/calcuis/gguf/releases))