Update README.md

README.md

@@ -6,7 +6,7 @@ datasets:
 
 ## Model Details
 
-This model is an int4 model(The vision module has also been quantized) with group_size 128 and symmetric quantization of [microsoft/Phi-3.5-vision-instruct](https://huggingface.co/microsoft/Phi-3.5-vision-instruct). Load the model with revision="" to use AutoGPTQ format.
+This model is an int4 model(The vision module has also been quantized) with group_size 128 and symmetric quantization of [microsoft/Phi-3.5-vision-instruct](https://huggingface.co/microsoft/Phi-3.5-vision-instruct).
 ## How To Use
 
 
@@ -40,13 +40,12 @@ model = AutoModelForCausalLM.from_pretrained(
   model_id, 
   device_map="auto", 
   trust_remote_code=True, 
-  torch_dtype="auto",
-  ##revision="" ##AutoGPTQ format
+  torch_dtype="auto"
 )
 processor = AutoProcessor.from_pretrained(model_id, 
   trust_remote_code=True, 
   num_crops=4
-) 
+)
 
 image_url = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg"
 content = "Describe this image."
@@ -114,11 +113,11 @@ auto-round-mllm --eval --model OPEA/Phi-3.5-vision-instruct-qvision-int4-sym-inc
 ```
 |Metric             |16bits|Pile Calib INT4  |
 |-------------------|:------|:------|
-|avg                |77.64 | |
-|MMBench_DEV_EN_V11 |71.83 | |
-|ScienceQA_VAL      |90.56 | |
-|TextVQA_VAL        |65.36 | |
-|POPE               |82.82 | |
+|avg                |77.64 |76.99 |
+|MMBench_DEV_EN_V11 |71.83 |71.05 |
+|ScienceQA_VAL      |90.56 |89.75 |
+|TextVQA_VAL        |65.36 |63.83 |
+|POPE               |82.82 |83.33 |
 
 ### Generate the model
 Here is the sample command to reproduce the model.