4 contributors

History: 78 commits

GiusFra

MI300 QKV fused and all layers sym, guidance scale 8, 10 steps, deterministic

ce39627 verified 4 months ago

all_linear_sym
QKV fused and all linear layers sym 4 months ago
all_linear_sym_7_5
QKV fused and all linear layers sym, guidance scale 7.5 4 months ago
all_linear_sym_7_5_calib12
MI250 QKV fused and all linear layers sym, guidance scale 7.5, calib steps 12 4 months ago
all_linear_sym_7_5_noconvout
MI250 QKV fused and all linear layers sym, guidance scale 7.5, no conv_out 4 months ago
full_sym
Full symmetric 4 months ago
fused_qkv
Fused QKV quant_params.json with zp 4 months ago
mi250_all_linear_sym_7_5
MI250 QKV fused and all linear layers sym, guidance scale 7.5 4 months ago
mi250_all_sym_8
MI250 QKV fused and all layers sym, guidance scale 8 4 months ago
mi250_all_sym_8_step10
MI250 QKV fused and all layers sym, guidance scale 8, 10 step, second try 4 months ago
mi250_all_sym_8_step12
MI250 Q 4 months ago
mi300_all_sym_8_step10
MI300 QKV fused and all layers sym, guidance scale 8, 10 step 4 months ago
mi300_all_sym_8_step10_det
MI300 QKV fused and all layers sym, guidance scale 8, 10 steps, deterministic 4 months ago
nvidia_all_sym_8
NVIDIA All sym, guidance scale 8 4 months ago
nvidia_all_sym_8_step10
NVIDIA All sym, guidance scale 8, step 10 4 months ago
qkv_sym
QKV fused and sym 4 months ago
quant_sdxl
Fix model loading 4 months ago
.gitattributes

1.63 kB

Updated quant_params 4 months ago
config.json

1.68 kB

Add config.json from stable-diffusion-xl-base-1.0/unet 5 months ago
math_model.py

9.05 kB

Update quant params structure (#2) 4 months ago
out.safetensors

7.11 GB
LFS

Output reference tensors 4 months ago
params.safetensors

5.14 GB
LFS

Updated params.safetensors 4 months ago
punet_inputs.safetensors

661 kB
LFS

Reference inputs 4 months ago
quant_param.json

85.1 MB
LFS

add missing smoothquant factors 5 months ago
quant_params.json

86.8 MB
LFS

Updated quant_params 4 months ago
test_quant_conv2d.py

1.18 kB

Update quant params structure (#2) 4 months ago
test_quant_linear.py

1.04 kB

Update quant params structure (#2) 4 months ago
vae.safetensors

167 MB
LFS

Added vae weights with FP16 fix. 4 months ago