{
  "schema": "computearena-public-benchmark/1",
  "submission_id": "1d2ef23a-606e-4c82-8a5b-9ba7218c055a",
  "run_id": "a5766a1c6785c835ab306b283223272e",
  "created_at_unix_ms": 1789845531292,
  "submitted_by": "prabod",
  "model": {
    "identity_resolution": "unresolved",
    "identity_verification": "unresolved",
    "name": "Qwen3-30B-A3B-Instruct-2507-cuda-q8",
    "architecture": "qwen3_moe",
    "quantization": "base_q4",
    "artifact": {
      "sha256": "36c5ed1a61e59449e23b5a4ec582dac844ec011efe91e6fb10e55a00380c9f53",
      "format": "base",
      "quantization_namespace": "basert",
      "quantization_scheme": "base_q4"
    }
  },
  "runtime": {
    "name": "basert",
    "version": "0.2.6"
  },
  "benchmark": {
    "mode": "text",
    "chip": "NVIDIA GB10",
    "backend": "cuda",
    "protocol": {
      "report_schema": "computearena-benchmark/1",
      "benchmark_schema": "basert-benchmark-harness/1",
      "telemetry_schema": "basert-telemetry/4",
      "throughput_schema": "computearena-throughput/3",
      "comparable": true,
      "context_isolation": "headline_then_per_prefill",
      "execution_layout": "headline_then_prefill_processes",
      "model_load_in_timing": false,
      "decode_initial_context_tokens": 1,
      "conditioning_schema": "basert-conditioning/1",
      "conditioning_mode": "warmup_only",
      "cooldown_enabled": false
    },
    "parameters": {
      "context_tokens": 16384,
      "decode_context_tokens": 1
    },
    "throughput": {
      "prefill": [
        {
          "tokens": 128,
          "tokens_per_second": 4115.178402753635
        },
        {
          "tokens": 256,
          "tokens_per_second": 6236.240580105138
        },
        {
          "tokens": 512,
          "tokens_per_second": 7253.006870553374
        },
        {
          "tokens": 1024,
          "tokens_per_second": 7544.838395651968
        },
        {
          "tokens": 2048,
          "tokens_per_second": 7183.623809569468
        },
        {
          "tokens": 4096,
          "tokens_per_second": 6304.864914285133
        },
        {
          "tokens": 8192,
          "tokens_per_second": 5361.256788922218
        },
        {
          "tokens": 16384,
          "tokens_per_second": 4135.056750940069
        }
      ],
      "decode": {
        "generated_tokens": 128,
        "tokens_per_second": 61.159759233107785
      }
    }
  }
}