{
  "schema": "computearena-public-benchmark/1",
  "submission_id": "bd8ef353-b49f-4edd-9262-a1609242b6e2",
  "run_id": "2b4e40409dc5c5957b767591e0e5193e",
  "created_at_unix_ms": 1789838790672,
  "submitted_by": "prabod",
  "model": {
    "canonical_id": "Qwen/Qwen3-4B",
    "identity_resolution": "artifact_verified",
    "identity_verification": "verified",
    "verified_at": "2026-09-11T12:18:35.929Z",
    "name": "Qwen3-4B-Q4",
    "architecture": "qwen",
    "quantization": "base_q4",
    "artifact": {
      "provider": "huggingface",
      "repo_id": "basecompute/Qwen3-4B",
      "revision": "7a98a7a34a8e0d1adf4a610fd8605b48d8beff77",
      "path": "Qwen3-4B-Q4.base",
      "sha256": "df660a56b61bb68b940da100f9ad345a93d396df1955387e2f41f317e130e696",
      "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": 2162.3558526603515
        },
        {
          "tokens": 256,
          "tokens_per_second": 2572.3929918290446
        },
        {
          "tokens": 512,
          "tokens_per_second": 2821.6644102790765
        },
        {
          "tokens": 1024,
          "tokens_per_second": 2848.0676246755284
        },
        {
          "tokens": 2048,
          "tokens_per_second": 2862.139537404699
        },
        {
          "tokens": 4096,
          "tokens_per_second": 2735.413925289881
        },
        {
          "tokens": 8192,
          "tokens_per_second": 2607.9478948493893
        },
        {
          "tokens": 16384,
          "tokens_per_second": 2333.2891074723116
        }
      ],
      "decode": {
        "generated_tokens": 128,
        "tokens_per_second": 91.34376620099745
      }
    }
  }
}