x

README.md

@@ -61,20 +61,21 @@
 - `api_key`：接口访问密钥（同时用于 8000 与 8001）
 - `tensor_parallel_size`：张量并行数，双卡建议 `2`
 - `dtype`：推理精度，默认 `bfloat16`
-- `selected_model`：当前生效模型，留空时回退到 `default_model`
+- `models.selected`：当前生效模型，留空时回退到 `models.default`
 
 ## config.json 说明
 
 `config.json` 采用以下结构：
 
-- 顶层 `default_model`：默认模型名
+- `models.default`：默认模型名
-- 顶层其他键：每个模型的配置档位
+- `models.selected`：当前生效模型名
+- `models.profiles`：模型配置集合
 - 每个模型至少建议包含：`hf_model_id`、`ctx`、`max_num_seqs`、`max_tokens`、`gpu_util`、`valid_tp`
 
 启动时会按以下优先级选模型：
 
-1. `config.json` 中 `selected_model`
+1. `config.json` 中 `models.selected`
-2. `config.json` 中 `default_model`
+2. `config.json` 中 `models.default`
 
 模型被选中后，会自动覆盖运行参数，包括：
 
@@ -88,7 +89,7 @@
 
 ## 部署步骤
 
-1. 修改 `config.json` 中的 `selected_model` 与服务参数。
+1. 修改 `config.json` 中的 `models.selected` 与服务参数。
 
 2. 构建并启动容器：
 
@@ -129,8 +130,9 @@ curl -X POST "http://localhost:8001/v1/chat/completions" \
 
 - Base URL 使用 `http://<服务器IP>:8001/v1`
 - API Key 使用 `config.json` 中 `api_key`
-- 模型名使用 `config.json` 中 `served_model_name`
+- 模型名使用 `config.json` 中 `models.profiles.<模型名>.served_model_name`
 - 若使用工具调用，`config.json` 中应配置 `tool_call_parser` 与 `enable_auto_tool_choice`
+- 所有路径按 Ubuntu 规范填写，本地模型建议使用 `/opt/model/<模型目录>`
 
 ## 双 AMD R9700 调优建议
 
@@ -138,4 +140,3 @@ curl -X POST "http://localhost:8001/v1/chat/completions" \
 - 首次部署建议设置 `GPU_MEMORY_UTILIZATION=0.90`，稳定后再调高
 - 若模型较大且吞吐压力高，可逐步调低 `MAX_MODEL_LEN` 或 `MAX_NUM_SEQS`
 - 确保宿主机已正确安装 ROCm 驱动并暴露 `/dev/kfd` 与 `/dev/dri`
-

config.json

@@ -9,62 +9,66 @@
       "port": 8001
     }
   },
-  "api_key": "",
+  "api_key": "sk-szcjw",
   "tensor_parallel_size": 2,
   "dtype": "bfloat16",
   "revision": "",
-  "default_model": "Qwen3.5-35B-A3B-GPTQ-Int4",
+  "models": {
-  "selected_model": "Qwen3.5-35B-A3B-GPTQ-Int4",
+    "default": "Qwen3.5-35B-A3B-GPTQ-Int4",
-  "Qwen3-Next-80B-A3B-Instruct-AWQ-4bit": {
+    "selected": "Qwen3.5-35B-A3B-GPTQ-Int4",
-    "ctx": "24576",
+    "profiles": {
-    "trust_remote": true,
+      "Qwen3-Next-80B-A3B-Instruct-AWQ-4bit": {
-    "valid_tp": [2],
+        "ctx": "24576",
-    "max_num_seqs": "32",
+        "trust_remote": true,
-    "max_tokens": "16384",
+        "valid_tp": [2],
-    "gpu_util": "0.98",
+        "max_num_seqs": "32",
-    "enforce_eager": false,
+        "max_tokens": "16384",
-    "env": {
+        "gpu_util": "0.98",
-      "VLLM_USE_TRITON_AWQ": "1"
+        "enforce_eager": false,
-    },
+        "env": {
-    "tool_call_parser": "qwen3_xml",
+          "VLLM_USE_TRITON_AWQ": "1"
-    "enable_auto_tool_choice": true,
+        },
-    "served_model_name": "Qwen3-Next-80B-A3B-Instruct-AWQ-4bit",
+        "tool_call_parser": "qwen3_xml",
-    "hf_model_id": "cpatonn/Qwen3-Next-80B-A3B-Instruct-AWQ-4bit"
+        "enable_auto_tool_choice": true,
-  },
+        "served_model_name": "Qwen3-Next-80B-A3B-Instruct-AWQ-4bit",
-  "GLM-4.7-Flash-AWQ": {
+        "hf_model_id": "cpatonn/Qwen3-Next-80B-A3B-Instruct-AWQ-4bit"
-    "ctx": "32768",
+      },
-    "trust_remote": true,
+      "GLM-4.7-Flash-AWQ": {
-    "valid_tp": [1, 2],
+        "ctx": "32768",
-    "max_num_seqs": "64",
+        "trust_remote": true,
-    "max_tokens": "32768",
+        "valid_tp": [1, 2],
-    "gpu_util": "0.98",
+        "max_num_seqs": "64",
-    "tool_call_parser": "qwen3_xml",
+        "max_tokens": "32768",
-    "enable_auto_tool_choice": true,
+        "gpu_util": "0.98",
-    "served_model_name": "GLM-4.7-Flash-AWQ",
+        "tool_call_parser": "qwen3_xml",
-    "hf_model_id": "THUDM/GLM-4.7-Flash-AWQ"
+        "enable_auto_tool_choice": true,
-  },
+        "served_model_name": "GLM-4.7-Flash-AWQ",
-  "Qwen3.5-27B-FP8": {
+        "hf_model_id": "THUDM/GLM-4.7-Flash-AWQ"
-    "ctx": "32768",
+      },
-    "trust_remote": true,
+      "Qwen3.5-27B-FP8": {
-    "valid_tp": [1, 2],
+        "ctx": "32768",
-    "max_num_seqs": "64",
+        "trust_remote": true,
-    "max_tokens": "32768",
+        "valid_tp": [1, 2],
-    "gpu_util": "0.98",
+        "max_num_seqs": "64",
-    "tool_call_parser": "qwen3_xml",
+        "max_tokens": "32768",
-    "enable_auto_tool_choice": true,
+        "gpu_util": "0.98",
-    "served_model_name": "Qwen3.5-27B-FP8",
+        "tool_call_parser": "qwen3_xml",
-    "hf_model_id": "RedHatAI/Qwen3.5-27B-FP8-dynamic"
+        "enable_auto_tool_choice": true,
-  },
+        "served_model_name": "Qwen3.5-27B-FP8",
-  "Qwen3.5-35B-A3B-GPTQ-Int4": {
+        "hf_model_id": "RedHatAI/Qwen3.5-27B-FP8-dynamic"
-    "ctx": "32768",
+      },
-    "trust_remote": true,
+      "Qwen3.5-35B-A3B-GPTQ-Int4": {
-    "valid_tp": [1, 2],
+        "ctx": "32768",
-    "max_num_seqs": "64",
+        "trust_remote": true,
-    "max_tokens": "32768",
+        "valid_tp": [1, 2],
-    "gpu_util": "0.98",
+        "max_num_seqs": "64",
-    "tool_call_parser": "qwen3_xml",
+        "max_tokens": "32768",
-    "enable_auto_tool_choice": true,
+        "gpu_util": "0.98",
-    "served_model_name": "Qwen3.5-35B-A3B-GPTQ-Int4",
+        "tool_call_parser": "qwen3_xml",
-    "hf_model_id": "Qwen/Qwen3.5-35B-A3B-GPTQ-Int4"
+        "enable_auto_tool_choice": true,
+        "served_model_name": "Qwen3.5-35B-A3B-GPTQ-Int4",
+        "hf_model_id": "Qwen/Qwen3.5-35B-A3B-GPTQ-Int4"
+      }
+    }
   }
 }

docker-compose.yml

@@ -10,7 +10,7 @@ services:
       - "8000:8000"
     volumes:
       - /opt/model:/opt/model:ro
-      - d:/project/rocm_vllm_nightly/rocm_vllm_nightly/config.json:/workspace/config.json:ro
+      - ./config.json:/workspace/config.json:ro
     devices:
       - /dev/kfd
       - /dev/dri
@@ -35,7 +35,7 @@ services:
       - "8001:8001"
     volumes:
       - /opt/model:/opt/model:ro
-      - d:/project/rocm_vllm_nightly/rocm_vllm_nightly/config.json:/workspace/config.json:ro
+      - ./config.json:/workspace/config.json:ro
     devices:
       - /dev/kfd
       - /dev/dri