x

config.json

@@ -22,28 +22,6 @@
     "default": "Qwen3.5-35B-A3B-GPTQ-Int4",
     "selected": "Qwen3.6-35B-A3B-FP8",
     "profiles": {
-      "Qwen3.6-35B-A3B": {
-        "local_path": "Qwen3.6-35B-A3B",
-        "dtype": "float16",
-        "quantization": "auto",
-        "ctx": "131072",
-        "max_tokens": "8192",
-        "trust_remote": true,
-        "enforce_eager": false,
-        "valid_tp": [
-          2
-        ],
-        "max_num_seqs": "12",
-        "gpu_util": "0.90",
-        "tool_call_parser": "qwen3_coder",
-        "reasoning_parser": "qwen3",
-        "enable_auto_tool_choice": true,
-        "served_model_name": "Qwen3.6-35B-A3B",
-        "hf_model_id": "Qwen/Qwen3.6-35B-A3B",
-        "env": {
-          "VLLM_RPC_TIMEOUT": "300"
-        }
-      },
       "Qwen3.6-35B-A3B-FP8": {
         "local_path": "Qwen3.6-35B-A3B-FP8",
         "dtype": "float16",

docker-compose.yml

@@ -3,6 +3,7 @@ services:
     build:
       context: .
       dockerfile: Dockerfile
+      pull: false
     image: vllm-openai-rocm:nightly
     container_name: rocm-vllm-openai
     entrypoint: ["python"]