Update README.md

README.md

@@ -24,13 +24,10 @@ base_model: THUDM/glm-4v-9b
 ## Usage
 This model is quantized using [AutoGPTQ](https://github.com/AutoGPTQ/AutoGPTQ) for [THUDM/glm-4v-9b](https://huggingface.co/THUDM/glm-4v-9b). 
 
-Use pip install AutoGPTQ (required)
+It is recommended to install AutoGPTQ by compiling from the source code. 
 
 (The quantization script will be released later)
 
-```bash
-pip install auto-gptq
-```
 
 Since the original auto-gptq library does not support the quantization of chatglm models, manual import (hack) is required.
 ```python