x

2026-04-01 01:08:27 +08:00
parent ba1d64f7a3
commit 1f807347ee
3 changed files with 17 additions and 5 deletions
@@ -73,18 +73,21 @@
        "local_path": "Qwen3.5-35B-A3B-GPTQ-Int4",
        "dtype": "float16",
        "quantization": "gptq",
-        "ctx": "131072",
+        "ctx": "65536",
        "trust_remote": true,
        "valid_tp": [1, 2],
-        "max_num_seqs": "8",
-        "max_tokens": "8192",
-        "gpu_util": "0.90",
+        "max_num_seqs": "4",
+        "max_tokens": "4096",
+        "gpu_util": "0.85",
        "enforce_eager": false,
        "tool_call_parser": "qwen3_coder",
        "reasoning_parser": "qwen3",
        "enable_auto_tool_choice": true,
        "served_model_name": "Qwen3.5-35B-A3B-GPTQ-Int4",
-        "hf_model_id": "Qwen/Qwen3.5-35B-A3B-GPTQ-Int4"
+        "hf_model_id": "Qwen/Qwen3.5-35B-A3B-GPTQ-Int4",
+        "env": {
+          "VLLM_RPC_TIMEOUT": "300"
+        }
      },
      "Qwen3.5-35B-A3B-FP8": {
        "local_path": "Qwen3.5-35B-A3B-FP8",