Hugging Face's logo

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

Safetensors
qwen2_vl
4-bit precision
intel/auto-round
Model card Files Community
cicdatopea commited on
Commit
a93058e
·
verified ·
1 Parent(s): 1e1482e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -111,7 +111,7 @@ messages = [
111
  ## 这张图片代表的是英特尔（Intel）公司。
112
 
113
  ```
114
- <!--
115
  ## Evaluation the model
116
  pip3 install git+https://github.com/open-compass/VLMEvalKit.git@7de2dcb. The evaluation process may encounter errors that require changing model backend or evaluation code. Detailed instructions will be provided in a future update.
117
  ```bash
@@ -119,11 +119,11 @@ auto-round-mllm --eval --model OPEA/Qwen2-VL-7B-Instruct-int4-sym-inc --tasks MM
119
  ```
120
  |Metric |16bits|Pile Calib INT4 |
121
  |:-------------------|:------|:------|
122
- |avg | | |
123
- |MMBench_DEV_EN_V11 | | |
124
- |ScienceQA_VAL | | |
125
- |TextVQA_VAL | | |
126
- |POPE | | | -->
127
 
128
  ### Generate the model
129
  Here is the sample command to reproduce the model.
 
111
  ## 这张图片代表的是英特尔（Intel）公司。
112
 
113
  ```
114
+
115
  ## Evaluation the model
116
  pip3 install git+https://github.com/open-compass/VLMEvalKit.git@7de2dcb. The evaluation process may encounter errors that require changing model backend or evaluation code. Detailed instructions will be provided in a future update.
117
  ```bash
 
119
  ```
120
  |Metric |16bits|Pile Calib INT4 |
121
  |:-------------------|:------|:------|
122
+ |avg |87.80 |87.63 |
123
+ |MMBench_DEV_EN_V11 |86.76 |86.30 |
124
+ |ScienceQA_VAL |91.65 |91.23 |
125
+ |TextVQA_VAL |85.45 |85.39 |
126
+ |POPE |87.32 |87.61 |
127
 
128
  ### Generate the model
129
  Here is the sample command to reproduce the model.