{
  "schema": "computearena-public-benchmark/1",
  "submission_id": "a8beff71-f75f-4363-be6a-2e92c03461dd",
  "run_id": "66dbf3d641eb5693b41f7d4f928338c6",
  "created_at_unix_ms": 1789843106354,
  "submitted_by": "prabod",
  "model": {
    "canonical_id": "meta-llama/Llama-3.2-3B-Instruct",
    "identity_resolution": "artifact_verified",
    "identity_verification": "verified",
    "verified_at": "2026-09-11T12:18:35.929Z",
    "name": "Llama-3.2-3B-Instruct-Q4",
    "architecture": "llama",
    "quantization": "base_q4",
    "artifact": {
      "provider": "huggingface",
      "repo_id": "basecompute/Llama-3.2-3B-Instruct",
      "revision": "c111e66fb8f31b78e374d67d8145504998787ea6",
      "path": "Llama-3.2-3B-Instruct-Q4.base",
      "sha256": "20c803b2eeee70a7dada67ca0ee2d75081d95d0e35bac27d1857f9e3fdd84615",
      "format": "base",
      "quantization_namespace": "basert",
      "quantization_scheme": "base_q4"
    }
  },
  "runtime": {
    "name": "basert",
    "version": "0.2.6"
  },
  "benchmark": {
    "mode": "text",
    "chip": "Apple M4 Max",
    "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": 1793.5072605833504
        },
        {
          "tokens": 256,
          "tokens_per_second": 2113.663267441008
        },
        {
          "tokens": 512,
          "tokens_per_second": 2201.60687593759
        },
        {
          "tokens": 1024,
          "tokens_per_second": 2252.0005229993913
        },
        {
          "tokens": 2048,
          "tokens_per_second": 2214.2772021487085
        },
        {
          "tokens": 4096,
          "tokens_per_second": 2137.1324359330906
        },
        {
          "tokens": 8192,
          "tokens_per_second": 1924.880453921666
        },
        {
          "tokens": 16384,
          "tokens_per_second": 1603.134146815764
        }
      ],
      "decode": {
        "generated_tokens": 128,
        "tokens_per_second": 178.7955647136174
      }
    }
  }
}