Bedovyy's picture
Update README.md
925ae0f verified
---
license: apache-2.0
base_model:
- Comfy-Org/ERNIE-Image
base_model_relation: quantized
tags:
- comfyui
- diffusion-single-file
pipeline_tag: text-to-image
---
# Quantized models of [ERNIE-Image](https://huggingface.co/baidu/ERNIE-Image) / [ERNIE-Image-Turbo](https://huggingface.co/baidu/ERNIE-Image-Turbo)
- nvfp4 (4.78GB)
- fp8e4m3 (8.22GB)
- int8rowwise (8.22GB) - it needs [ComfyUI-INT8-Fast](https://github.com/BobJohnson24/ComfyUI-INT8-Fast) custom node
## Generation Speed
### ERINE-Image-Turbo
| GPU | Quantization | Speed (it/s) | Time (secs) | vs BF16 |
|----------|--------------|--------------|-------------|---------------------|
| RTX 5090 | bf16 | 2.09 | 4.87 | 100% |
| | fp8e4m3 | 3.69 | 3.32 | 147% |
| | int8rowwise | 4.31 | 3.05 | *160%* |
| | nvfp4 | 5.09 | 2.72 | **179%** |
| RTX 3090 | bf16 | 0.88 | 12.42 | *100%* |
| | fp8e4m3 | 0.84 | 12.73 | 98% |
| | int8rowwise | 1.66 | 7.04 | **176%** |
| | nvfp4 | 0.83 | 12.71 | 98% |
| RTX 3060 | bf16 | 0.26 | 43.02 | 100% |
| | fp8e4m3 | 0.39 | 28.66 | *150%* |
| | int8rowwise | 0.82 | 14.43 | **298%** |
| | nvfp4 | 0.39 | 28.72 | 150% |
### ERINE-Image
| GPU | Quantization | Speed (it/s) | Time (secs) | vs BF16 |
|----------|--------------|--------------|-------------|---------------------|
| RTX 5090 | bf16 | 1.08 | 20.08 | 100% |
| | fp8e4m3 | 1.97 | 11.67 | 172% |
| | int8rowwise | 2.14 | 10.89 | *184%* |
| | nvfp4 | 2.56 | 9.35 | **215%** |
| RTX 3090 | bf16 | 0.40 | 53.33 | *100%* |
| | fp8e4m3 | 0.39 | 54.71 | 97% |
| | int8rowwise | 0.79 | 28.08 | **190%** |
| | nvfp4 | 0.38 | 55.20 | 97% |
| RTX 3060 | bf16 | 0.11 | 201.41 | 100% |
| | fp8e4m3 | 0.17 | 130.48 | *154%* |
| | int8rowwise | 0.35 | 62.42 | **323%** |
| | nvfp4 | 0.17 | 130.87 | 154% |
## Sample
### ERNIE-Image-Turbo
![ernie-image-turbo-comp1](https://cdn-uploads.huggingface.co/production/uploads/63fbf6951b4b1bd4e706fed1/14jzEohjbhZfR7NNFb6OT.webp)
![ernie-image-turbo-comp2](https://cdn-uploads.huggingface.co/production/uploads/63fbf6951b4b1bd4e706fed1/DbdZqEd8ggd7UNBNBoGBD.webp)
![ernie-image-turbo-comp3](https://cdn-uploads.huggingface.co/production/uploads/63fbf6951b4b1bd4e706fed1/QARRTSNDpe-4f4DAjX-_2.webp)
![erine-image-turbo-comp4](https://cdn-uploads.huggingface.co/production/uploads/63fbf6951b4b1bd4e706fed1/lz3daUybkP-ggSNtV0Nkv.webp)
### ERNIE-Image
![ernie-image-comp1](https://cdn-uploads.huggingface.co/production/uploads/63fbf6951b4b1bd4e706fed1/Hy24cvgMbDFsvTgQeXw5Z.webp)
![ernie-image-comp2](https://cdn-uploads.huggingface.co/production/uploads/63fbf6951b4b1bd4e706fed1/Ro03949AktXAEqzJcmder.webp)
![ernie-image-comp3](https://cdn-uploads.huggingface.co/production/uploads/63fbf6951b4b1bd4e706fed1/hN0Ow5MRznmTONz20Nyr5.webp)
![erine-image-comp4](https://cdn-uploads.huggingface.co/production/uploads/63fbf6951b4b1bd4e706fed1/6ZZ3_Cj1brN3cQ4dDPpGv.webp)
## How to reproduce
Use https://github.com/bedovyy/comfy-dit-quantizer with the below config json.
```json
{
"block_names": ["layers"],
"rules": [
{ "policy": "keep", "match": ["adaLN", "self_attention.norm"] },
{ "policy": "float8_e4m3fn", "match": ["mlp", "self_attention.to"] }
]
}
```