amd-shark
/

sdxl-quant-fp8

Model card Files Files and versions

60.6 GB

Ctrl+K

Ctrl+K

4 contributors

History: 36 commits

GiusFra's picture

Upload unet_int8_sdpa_fp8_vae_int8/vae_quant_params.json with huggingface_hub

f61f04f verified about 1 year ago

all_linear_sym_8_calib8
Fix names almost 2 years ago
all_quant_int8_sdpa_fp8
Upload all_quant_int8_sdpa_fp8/params.safetensors with huggingface_hub about 1 year ago
all_sym_8_calib10
MI250 QKV fused and all layers sym, FP8 attention, guidance scale 8, calib steps 10 almost 2 years ago
brevitas
updated quant_params with QKV fusion almost 2 years ago
linear_conv_fp8_sdpa_fp16_eq_bl
Create config.json over 1 year ago
linear_conv_fp8_sdpa_fp16_no_eq_bl
Create config.json over 1 year ago
linear_conv_fp8_sdpa_fp8_eq_bl
Create config.json over 1 year ago
linear_conv_fp8_sdpa_fp8_no_eq_bl
Create config.json over 1 year ago
nvidia_fp8_unet
Upload nvidia_fp8_unet/params.safetensors with huggingface_hub over 1 year ago
unet_int8_sdpa_fp8_vae_int8
Upload unet_int8_sdpa_fp8_vae_int8/vae_quant_params.json with huggingface_hub about 1 year ago
.gitattributes

2.25 kB
Upload unet_int8_sdpa_fp8_vae_int8/unet_quant_params.json with huggingface_hub about 1 year ago
attn.py

6.26 kB
Added SDPA math model & test over 1 year ago
math_model.py

7.13 kB
Create math_model.py over 1 year ago
sdxl.json

2.19 MB
Upload sdxl.json with huggingface_hub almost 2 years ago
sdxl.safetensors

5.14 GB
xet

Upload sdxl.safetensors with huggingface_hub almost 2 years ago
test_attn.py

1.31 kB
[math_model] Make it more obvious that softmax scale comes from the quantizer over 1 year ago