{
  "engine": "vLLM",
  "engine_version": "0.24.0",
  "checkpoint_precision": "bfloat16",
  "evaluation_quantization": "fp8 on the fly from bfloat16",
  "maximum_model_length": 262144,
  "tensor_parallel_size": 1,
  "gpu_memory_utilization": 0.92,
  "maximum_sequences": 1,
  "maximum_batched_tokens": 16384,
  "kv_cache_dtype": "auto",
  "prefix_caching": true,
  "chunked_prefill": true,
  "maximum_logprobs": 5,
  "language_model_only": true,
  "generation_config_source": "vllm",
  "reasoning_parser": "qwen3",
  "deep_gemm_enabled": false,
  "renderer": "qwen3_5",
  "thinking": {
    "enabled": true,
    "preserved_in_history": true
  },
  "response_format": "json_schema",
  "provider_fallback_allowed": false,
  "maximum_total_action_samples": 4,
  "timeout_seconds": 3600
}