在model_config.py.example中增加qwen量化模型启动的说明

2023-12-15 14:23:34 +08:00 · 2023-12-15 14:23:34 +08:00 · a870076051
parent 7e01e82470
commit a870076051
1 changed files with 2 additions and 0 deletions
--- a/configs/model_config.py.example
+++ b/configs/model_config.py.example
@ -179,6 +179,8 @@ MODEL_PATH = {

        "Qwen-14B": "Qwen/Qwen-14B",
        "Qwen-14B-Chat": "Qwen/Qwen-14B-Chat",
+        # 在新版的transformers下需要手动修改模型的config.json文件，在quantization_config字典中
+        # 增加`disable_exllama:true` 字段才能启动qwen的量化模型
        "Qwen-14B-Chat-Int8": "Qwen/Qwen-14B-Chat-Int8",
        "Qwen-14B-Chat-Int4": "Qwen/Qwen-14B-Chat-Int4",