head bits 8 default

doesn't make much of a vram difference for better results

auto-exl2-upload/auto-exl2-upload.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:180db595eefd1f31ec9b985db95c5f3326a79ad91329ba02c7b4d6676015f4d4
-size 7219
+oid sha256:46c9216642f00007b56581890c3ca89426a6ce7148bbca4f19a5ec0a04ac5e16
+size 7202

auto-exl2-upload/exl2-quant.py

@@ -168,7 +168,7 @@ for bpw in bpwvalue:
     os.makedirs(f"{model}-exl2-{bpw}bpw", exist_ok=True) #create compile full directory
     subprocess.run(f"{oscp} models{slsh}{model}{slsh}config.json {model}-exl2-{bpw}bpw-WD", shell=True) #copy config to working directory
     #more settings exist in the convert.py script, to veiw them go to docs/convert.md or https://github.com/turboderp/exllamav2/blob/master/doc/convert.md
-    result = subprocess.run(f"{pyt} exllamav2/convert.py -i models/{model} -o {model}-exl2-{bpw}bpw-WD -cf {model}-exl2-{bpw}bpw -b {bpw}{mskip}", shell=True) #run quantization and exit if failed (Credit to turbo for his dedication to exl2)
+    result = subprocess.run(f"{pyt} exllamav2/convert.py -i models/{model} -o {model}-exl2-{bpw}bpw-WD -cf {model}-exl2-{bpw}bpw -b {bpw}{mskip} -hb 8", shell=True) #run quantization and exit if failed (Credit to turbo for his dedication to exl2)
     if result.returncode != 0:
         print("Quantization failed.")
         sys.exit("Exiting...")

exl2-multi-quant-local/exl2-multi-quant-local.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bae7404db9d1534b0eec7a865cf53dc17b29111e7107e152bde428ceb819e834
-size 6229
+oid sha256:030dc763b103c4d58919a31ed238ff839bd0b1b0ec5dd00a0bcdd1571eed778b
+size 6211

exl2-multi-quant-local/exl2-quant.py

@@ -122,7 +122,7 @@ for bpw in bpwvalue:
     os.makedirs(f"{modelname}-exl2-quants{slsh}{modelname}-exl2-{bpw}bpw", exist_ok=True) #create compile full directory
     subprocess.run(f"{oscp} models{slsh}{model}{slsh}config.json {model}-exl2-{bpw}bpw-WD", shell=True) #copy config to working directory
     #more settings exist in the convert.py script, to veiw them go to docs/convert.md or https://github.com/turboderp/exllamav2/blob/master/doc/convert.md
-    result = subprocess.run(f"{pyt} exllamav2/convert.py -i models/{model} -o {model}-exl2-{bpw}bpw-WD -cf {modelname}-exl2-quants{slsh}{modelname}-exl2-{bpw}bpw -b {bpw}{mskip}", shell=True) #run quantization and exit if failed (Credit to turbo for his dedication to exl2)
+    result = subprocess.run(f"{pyt} exllamav2/convert.py -i models/{model} -o {model}-exl2-{bpw}bpw-WD -cf {modelname}-exl2-quants{slsh}{modelname}-exl2-{bpw}bpw -b {bpw}{mskip} -hb 8", shell=True) #run quantization and exit if failed (Credit to turbo for his dedication to exl2)
     if result.returncode != 0:
         print("Quantization failed.")
         sys.exit("Exiting...")