Upload folder using huggingface_hub

quanto-exps-2/ckpt@auraflow-bs@1-dtype@bf16-qtype@int8-qte@0-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 1, "memory": "9.917", "time": "13.604", "dtype": "bf16", "qtype": "int8", "qte": 0, "exclude_layers": null}

quanto-exps-2/ckpt@auraflow-bs@1-dtype@bf16-qtype@int8-qte@1-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 1, "memory": "8.840", "time": "13.628", "dtype": "bf16", "qtype": "int8", "qte": 1, "exclude_layers": null}

quanto-exps-2/ckpt@auraflow-bs@1-dtype@fp16-qtype@int8-qte@0-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 1, "memory": "10.520", "time": "13.892", "dtype": "fp16", "qtype": "int8", "qte": 0, "exclude_layers": null}

quanto-exps-2/ckpt@auraflow-bs@1-dtype@fp16-qtype@int8-qte@1-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 1, "memory": "8.975", "time": "13.949", "dtype": "fp16", "qtype": "int8", "qte": 1, "exclude_layers": null}

quanto-exps-2/ckpt@auraflow-bs@4-dtype@bf16-qtype@int8-qte@0-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 4, "memory": "9.915", "time": "49.508", "dtype": "bf16", "qtype": "int8", "qte": 0, "exclude_layers": null}

quanto-exps-2/ckpt@auraflow-bs@4-dtype@bf16-qtype@int8-qte@1-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 4, "memory": "8.839", "time": "49.459", "dtype": "bf16", "qtype": "int8", "qte": 1, "exclude_layers": null}

quanto-exps-2/ckpt@auraflow-bs@4-dtype@fp16-qtype@int8-qte@0-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 4, "memory": "10.520", "time": "51.264", "dtype": "fp16", "qtype": "int8", "qte": 0, "exclude_layers": null}

quanto-exps-2/ckpt@auraflow-bs@4-dtype@fp16-qtype@int8-qte@1-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 4, "memory": "8.973", "time": "51.239", "dtype": "fp16", "qtype": "int8", "qte": 1, "exclude_layers": null}

quanto-exps-2/ckpt@pixart-bs@1-dtype@bf16-qtype@int8-qte@0-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 1, "memory": "9.847", "time": "1.402", "dtype": "bf16", "qtype": "int8", "qte": 0, "exclude_layers": null}

quanto-exps-2/ckpt@pixart-bs@1-dtype@bf16-qtype@int8-qte@1-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 1, "memory": "5.537", "time": "1.422", "dtype": "bf16", "qtype": "int8", "qte": 1, "exclude_layers": null}

quanto-exps-2/ckpt@pixart-bs@1-dtype@fp16-qtype@int8-qte@0-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 1, "memory": "11.722", "time": "1.456", "dtype": "fp16", "qtype": "int8", "qte": 0, "exclude_layers": null}

quanto-exps-2/ckpt@pixart-bs@1-dtype@fp16-qtype@int8-qte@1-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 1, "memory": "5.538", "time": "1.494", "dtype": "fp16", "qtype": "int8", "qte": 1, "exclude_layers": null}

quanto-exps-2/ckpt@pixart-bs@4-dtype@bf16-qtype@int8-qte@0-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 4, "memory": "9.848", "time": "4.924", "dtype": "bf16", "qtype": "int8", "qte": 0, "exclude_layers": null}

quanto-exps-2/ckpt@pixart-bs@4-dtype@bf16-qtype@int8-qte@1-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 4, "memory": "5.538", "time": "4.933", "dtype": "bf16", "qtype": "int8", "qte": 1, "exclude_layers": null}

quanto-exps-2/ckpt@pixart-bs@4-dtype@fp16-qtype@int8-qte@0-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 4, "memory": "11.723", "time": "4.959", "dtype": "fp16", "qtype": "int8", "qte": 0, "exclude_layers": null}

quanto-exps-2/ckpt@pixart-bs@4-dtype@fp16-qtype@int8-qte@1-fuse@1_info.json

@@ -0,0 +1 @@
+{"batch_size": 4, "memory": "5.538", "time": "5.003", "dtype": "fp16", "qtype": "int8", "qte": 1, "exclude_layers": null}

quanto-exps-2/launch_exps_2.sh

@@ -1,10 +1,10 @@
 #!/bin/bash
 
 # Possible values for each argument
-ckpt_ids=("stabilityai/stable-diffusion-3-medium-diffusers" "PixArt-alpha/PixArt-Sigma-XL-2-1024-MS" "fal/AuraFlow")
+ckpt_ids=("PixArt-alpha/PixArt-Sigma-XL-2-1024-MS" "fal/AuraFlow")
 batch_sizes=(1 4)
 torch_dtypes=("fp16" "bf16")
-qtypes=("int4")
+qtypes=("int8")
 qte=(0 1)
 
 # Loop over all combinations
@@ -21,7 +21,7 @@ for ckpt_id in "${ckpt_ids[@]}"; do
           fi
 
           # Run the Python script with the current combination of arguments
-          python3 benchmark.py --ckpt_id $ckpt_id --batch_size $batch_size --torch_dtype $torch_dtype $q_arg --qte $te --exclude_layers=proj_out
+          python3 benchmark.py --ckpt_id $ckpt_id --batch_size $batch_size --torch_dtype $torch_dtype $q_arg --qte $te --fuse 1
         done
       done
     done