| --- |
| license: other |
| license_name: ideogram-4-non-commercial |
| license_link: https://huggingface.co/ideogram-ai/ideogram-4-fp8/blob/main/LICENSE.md |
| pipeline_tag: text-to-image |
| tags: |
| - text-to-image |
| - image-generation |
| - diffusion |
| - flow-matching |
| - dit |
| - ideogram |
| --- |
| |
| This is an unavoidable double quantization due to the release state of Ideogram4. |
|
|
| The FP8 weights were cast to FP32 with the FP8 scales, then downcast to BF16 before being converted to INT8. |
|
|
| For use in ComfyUI with https://github.com/BobJohnson24/ComfyUI-INT8-Fast |
|
|
| Speed is 1.78x faster(2.03s/it) than FP8(3.62s/it) on my 3090, without compile. |
|
|
| <s>~2x faster with torch compile.</s> |
|
|
| After further inspection, it appears there may be quality issues with torch compiling this model. |
|
|
| Quick comparison: |
|
|
| <img src="Comparison.jpg" width="1000" height="500"> |
|
|
| <img src="Comparison2.jpg" width="1000" height="500"> |
