feat(*):增加了一些模型

vllm-qwen25-coder-32b-instruct/compose.yaml

@@ -18,8 +18,8 @@ services:
     runtime: nvidia
     volumes:
       - /home/deepgeek/data/data_local/server/vllm/models:/models
-    restart: always
-    command: --served-model-name Qwen2.5-Coder-32B-Instruct --model
+    restart: unless-stopped
+    command: --served-model-name Qwen2.5-Coder:32b --model
       /models/Qwen/Qwen2.5-Coder-32B-Instruct --trust-remote-code --host 0.0.0.0
       --port 8080 --max-model-len 8192 --tensor-parallel-size 2
       --gpu_memory_utilization 0.9 --enforce-eager --dtype auto --swap-space 8