Text-to-Image
Diffusers
Safetensors
English
Chinese
QwenImage21Pipeline
bitsandbytes
int8
image-generation
image-editing
rgba
8-bit precision
Instructions to use ixim/Image21-INT8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use ixim/Image21-INT8 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("ixim/Image21-INT8", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Download evaluation/comparison.csv from ixim/Image21-INT8: direct link, hf CLI and curl.
- Browser
- Download file 2.49 kB
-
https://huggingface.co/ixim/Image21-INT8/resolve/main/evaluation/comparison.csv
- Command line
-
hf download hf://ixim/Image21-INT8/evaluation/comparison.csv
-
curl -L -o comparison.csv https://huggingface.co/ixim/Image21-INT8/resolve/main/evaluation/comparison.csv
2.49 kB
| case_id,seed,bf16_seconds,int8_seconds,speed_ratio_bf16_over_int8,bf16_peak_allocated_gib,int8_peak_allocated_gib,rgb_mae_white,rgb_psnr_white_db,alpha_mae | |
| portrait,42,28.585568599999533,19.621747900000628,1.4568308973125996,16.40157413482666,9.490443229675293,0.013405879028141499,31.702880072340637,4.5881097321398556e-05 | |
| portrait,123,26.070829300000696,19.478468499999508,1.3384434869713373,16.40157413482666,9.490443229675293,0.0162954181432724,30.546014578484698,0.0001803267077775672 | |
| english_text,42,25.41653260000021,19.543290399999023,1.300524736612494,16.409088134765625,9.499232769012451,0.026258574798703194,19.447587010373717,8.03704188001575e-06 | |
| english_text,123,25.66373449999992,19.677271200000177,1.3042323927516783,16.409088134765625,9.499234199523926,0.011798395775258541,25.825570953357694,1.0239845323667396e-05 | |
| chinese_text,42,25.618986999999834,19.60913740000069,1.30648209951341,16.413254261016846,9.495181560516357,0.05972025915980339,15.60138057172976,0.0001082326634787023 | |
| chinese_text,123,25.45504400000027,19.6146719999997,1.2977552721758825,16.413254261016846,9.495178699493408,0.024602750316262245,20.44133777040833,6.83055113768205e-05 | |
| composition,42,25.39061010000114,19.315522199998668,1.3145184394757337,16.411731719970703,9.494652271270752,0.02268194407224655,23.666249907034246,2.1964429834042676e-05 | |
| composition,123,25.480030999999144,19.14716750000116,1.3307467540563167,16.411731719970703,9.494656562805176,0.01959226094186306,22.865933886645216,3.4590328141348436e-05 | |
| texture,42,25.31820230000085,19.42428639999889,1.3034302408145246,16.39810037612915,9.48720407485962,0.06978315860033035,18.01562989780135,0.00019098912889603525 | |
| texture,123,25.184098900001118,19.159069299999828,1.314474023015374,16.39810037612915,9.487202167510986,0.01683949865400791,27.92905075564026,5.856681673321873e-05 | |
| rgba,42,25.218287199999395,19.219452700001057,1.3121230658143563,16.395768642425537,9.477011680603027,0.018291756510734558,20.893335328149618,0.01108772773295641 | |
| rgba,123,25.541228100000808,19.219784199998685,1.3289029592747745,16.395768642425537,9.477009296417236,0.04646634683012962,17.051910336330792,0.029543064534664154 | |
| edit,42,30.789263599999686,25.072887500000434,1.2279903381690342,19.081242561340332,12.087220668792725,0.009711389429867268,36.48036013354599,0.00011113862274214625 | |
| edit,123,30.458208000000013,24.810640600000625,1.2276268271766932,19.081242561340332,12.087220668792725,0.003908749669790268,43.267885143031855,2.616058372950647e-05 | |