Hugging Face's logo

amd-shark
/
sdxl-quant-fp8

Model card Files
xet
Community
1
sdxl-quant-fp8

Commit History

Upload quant_sdxl_exps/sdxl_quant_artifacts_mxfp4_net2/net_2_mxfp4.irpa with huggingface_hub
e144c20
verified

GiusFra commited on

Upload quant_sdxl_exps/sdxl_quant_artifacts_mxfp4_net2/net_2_mxfp4.irpa with huggingface_hub
b194a28
verified

GiusFra commited on

Upload fp8_irpa/unet_quant_bias.irpa with huggingface_hub
b77d9e8
verified

GiusFra commited on

Upload fp8_irpa/unet.irpa with huggingface_hub
8ba4744
verified

GiusFra commited on

Upload fp8_irpa/unet.irpa with huggingface_hub
706b6f1
verified

GiusFra commited on

Upload folder using huggingface_hub
2b53661
verified

GiusFra commited on

Upload folder using huggingface_hub
74940ec
verified

GiusFra commited on

Upload folder using huggingface_hub
e020b38
verified

GiusFra commited on

Upload folder using huggingface_hub
189838c
verified

GiusFra commited on

Upload fp8_irpa/unet.irpa with huggingface_hub
2da830c
verified

GiusFra commited on

Upload fp8_irpa/unet.irpa with huggingface_hub
c544312
verified

GiusFra commited on

Upload fp8_irpa/unet.irpa with huggingface_hub
b60be3e
verified

GiusFra commited on

Upload fp8_irpa/unet.irpa with huggingface_hub
c6311dd
verified

GiusFra commited on

Upload fp8_irpa/unet.irpa with huggingface_hub
993b772
verified

GiusFra commited on

Upload fp8_irpa/unet.irpa with huggingface_hub
bd4c505
verified

GiusFra commited on

Upload fp8_irpa/unet.irpa with huggingface_hub
84258f6
verified

GiusFra commited on

Upload fp8_irpa/unet.irpa with huggingface_hub
d635292
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_ocp/params.safetensors with huggingface_hub
e6e3c03
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_ocp/quant_params.json with huggingface_hub
832910d
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8_v2/vae_quant_params.json with huggingface_hub
9436fb6
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8_v2/vae_params.safetensors with huggingface_hub
f4b3910
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8_v2/vae_params.safetensors with huggingface_hub
b1a165a
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8_v2/params.safetensors with huggingface_hub
7d0b300
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8_v2/quant_params.json with huggingface_hub
75d97e8
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8_v2/vae_params.safetensors with huggingface_hub
2266191
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8_v2/vae_quant_params.json with huggingface_hub
bb58fb1
verified

GiusFra commited on

Create config.json
ae57958
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8/vae_quant_params.json with huggingface_hub
f61f04f
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8/unet_quant_params.json with huggingface_hub
59590aa
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8/vae_params.safetensors with huggingface_hub
1dbb8b4
verified

GiusFra commited on

Upload unet_int8_sdpa_fp8_vae_int8/unet_params.safetensors with huggingface_hub
99cda0b
verified

GiusFra commited on

Upload all_quant_int8_sdpa_fp8/params.safetensors with huggingface_hub
8e60988
verified

GiusFra commited on

Upload all_quant_int8_sdpa_fp8/quant_params.json with huggingface_hub
008bca6
verified

GiusFra commited on

[math_model] Make it more obvious that softmax scale comes from the quantizer
db5a15b

nickfraser commited on

Create math_model.py
6f59b43
verified

GiusFra commited on

Upload nvidia_fp8_unet/params.safetensors with huggingface_hub
d9e66a0
verified

GiusFra commited on

Upload nvidia_fp8_unet/quant_params.json with huggingface_hub
730c8f5
verified

GiusFra commited on

Upload nvidia_fp8_unet/results_mlperf.json with huggingface_hub
f4037ed
verified

GiusFra commited on

Upload nvidia_fp8_unet/args.json with huggingface_hub
4e70299
verified

GiusFra commited on

Create config.json
b0f9624
verified

GiusFra commited on

Create config.json
b7db598
verified

GiusFra commited on

Create config.json
864a3a2
verified

GiusFra commited on

Create config.json
25e566b
verified

GiusFra commited on

Updated sdpa fp8 models
fa0155f

nickfraser commited on

Added models that are fully quantized with FP8.
cfd94d7

nickfraser commited on

Added SDPA math model & test
3fea540

nickfraser commited on

Fix names
740d40f

GiusFra commited on

MI250 QKV fused and all linear layers sym, FP8 attention, guidance scale 8, calib steps 8
b8d5ec9
verified

GiusFra commited on

Fix names
08a2fb9

GiusFra commited on

MI250 QKV fused and all linear layers sym, FP8 attention, guidance scale 8, calib steps 10
7c9637e
verified

GiusFra commited on