OPEA
/

Qwen2-VL-72B-Instruct-int4-sym-inc

4-bit precision

Model card Files Files and versions

cicdatopea commited on Jun 5

Commit

d01b175

·

verified ·

1 Parent(s): 474480b

Update config.json

Files changed (1) hide show

config.json +2 -2

config.json CHANGED Viewed

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/data5/models//Qwen2-VL-72B-Instruct",
   "architectures": [
     "Qwen2VLForConditionalGeneration"
   ],
@@ -20,7 +20,7 @@
   "quantization_config": {
     "amp": true,
     "autoround_version": "0.5.1",
-    "backend": "auto_round:gptq:exllamav2",
     "batch_size": 8,
     "bits": 4,
     "data_type": "int",

 {
+  "_name_or_path": "/data5/models/Qwen2-VL-72B-Instruct",
   "architectures": [
     "Qwen2VLForConditionalGeneration"
   ],
   "quantization_config": {
     "amp": true,
     "autoround_version": "0.5.1",
+    "packing_format": "auto_round:auto_gptq",
     "batch_size": 8,
     "bits": 4,
     "data_type": "int",