feature: add

README.md

@@ -19,7 +19,7 @@ docker pull harbor.4pd.io/mic-llm-x/combricon_vllm_mlu_wyq:1.0.0
 3. 启动docker 
 ```bash
 docker run -it --rm \
-  -p 100086:8000 \
+  -p 10086:8000 \
   -e MLU_VISIBLE_DEVICES=6 \
   --name test_wyq1 \
   -v /mnt/contest_ceph/wenyunqing/models/Qwen1.5-1.8B-Chat:/model:rw \
@@ -31,7 +31,7 @@ docker run -it --rm \
 注意需要在本地使用寒武纪 mlu370 芯片   
 4. 测试服务  
 ```bash
-curl http://localhost:80/v1/chat/completions \
+curl http://localhost: 10086/v1/chat/completions \
   -H "Content-Type: application/json" \
   -d '{
     "model": "llm",