ussoewwin commited on
Commit
cb199d4
·
verified ·
1 Parent(s): 4e8a038

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -1
README.md CHANGED
@@ -37,7 +37,8 @@ High-fidelity FP8 quantization for diffusion models (Z Image). HSWQ uses **sensi
37
 
38
  | Feature | V1: Standard Compatible | V2: High Performance Scaled |
39
  | :--- | :--- | :--- |
40
- | **Compatibility** | Full (100%), any FP8 loader | Custom loader (HSWQLoader) required |
 
41
  | **File format** | Standard FP8 (`torch.float8_e4m3fn`) | Extended FP8 (weights + `.scale` metadata) |
42
  | **Image quality (SSIM)** | ~0.89 (theoretical limit) | ~Unable to measure at this time |
43
  | **Mechanism** | Optimal clipping (smart clipping) | Full-range scaling (dynamic scaling) |
 
37
 
38
  | Feature | V1: Standard Compatible | V2: High Performance Scaled |
39
  | :--- | :--- | :--- |
40
+ | **Compatibility** | Full (100%), any FP8 loader | The scaled model does not perform well in the current ComfyUI.
41
+ |
42
  | **File format** | Standard FP8 (`torch.float8_e4m3fn`) | Extended FP8 (weights + `.scale` metadata) |
43
  | **Image quality (SSIM)** | ~0.89 (theoretical limit) | ~Unable to measure at this time |
44
  | **Mechanism** | Optimal clipping (smart clipping) | Full-range scaling (dynamic scaling) |