{
  "schema": "computearena-public-benchmark/1",
  "submission_id": "6ba444fd-e1aa-4df9-8628-317e5acfa930",
  "run_id": "4ad52732d6aeb3484dcdb172639b55db",
  "created_at_unix_ms": 1789881416924,
  "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-Q8",
    "architecture": "qwen",
    "quantization": "base_q8",
    "artifact": {
      "provider": "huggingface",
      "repo_id": "basecompute/Qwen3-4B",
      "revision": "7a98a7a34a8e0d1adf4a610fd8605b48d8beff77",
      "path": "Qwen3-4B-Q8.base",
      "sha256": "7898a470ac96c67b97e6330eb0f2b5fc00475247880a66e12b7640d9cc169b9a",
      "format": "base",
      "quantization_namespace": "basert",
      "quantization_scheme": "base_q8"
    }
  },
  "runtime": {
    "name": "basert",
    "version": "0.2.6"
  },
  "benchmark": {
    "mode": "text",
    "chip": "Apple M3 Ultra",
    "backend": "metal",
    "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": 1756.7894137471192
        },
        {
          "tokens": 256,
          "tokens_per_second": 2176.1015041774876
        },
        {
          "tokens": 512,
          "tokens_per_second": 2441.0884447412195
        },
        {
          "tokens": 1024,
          "tokens_per_second": 2737.3562868564127
        },
        {
          "tokens": 2048,
          "tokens_per_second": 2676.611473420177
        },
        {
          "tokens": 4096,
          "tokens_per_second": 2622.3840887822457
        },
        {
          "tokens": 8192,
          "tokens_per_second": 2287.9775980075533
        },
        {
          "tokens": 16384,
          "tokens_per_second": 1817.0363283484958
        }
      ],
      "decode": {
        "generated_tokens": 128,
        "tokens_per_second": 116.84708387757117
      }
    }
  }
}
