{
  "schema": "computearena-public-benchmark/1",
  "submission_id": "35fa5362-9ec8-4807-b144-70662efb60ff",
  "run_id": "0455ec1b5d0134428caaf1682fcaad14",
  "created_at_unix_ms": 1789839869970,
  "submitted_by": "prabod",
  "model": {
    "canonical_id": "meta-llama/Llama-3.1-8B-Instruct",
    "identity_resolution": "artifact_verified",
    "identity_verification": "verified",
    "verified_at": "2026-09-11T12:18:35.929Z",
    "name": "Llama-3.1-8B-Instruct-Q8",
    "architecture": "llama",
    "quantization": "base_q4",
    "artifact": {
      "provider": "huggingface",
      "repo_id": "basecompute/Llama-3.1-8B-Instruct",
      "revision": "d0f2c4c5af5c865e6e6427397ec6b7766bdedb64",
      "path": "Llama-3.1-8B-Instruct-Q8.base",
      "sha256": "833e8c57398d0878c7e2684bcb570dc9c3d0b5a9eb6107fa337e631b50af6f87",
      "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": 2845.259648002662
        },
        {
          "tokens": 256,
          "tokens_per_second": 4974.162337500664
        },
        {
          "tokens": 512,
          "tokens_per_second": 7298.8954309890205
        },
        {
          "tokens": 1024,
          "tokens_per_second": 8070.868626445109
        },
        {
          "tokens": 2048,
          "tokens_per_second": 7919.87029063731
        },
        {
          "tokens": 4096,
          "tokens_per_second": 7338.68898459068
        },
        {
          "tokens": 8192,
          "tokens_per_second": 6647.205121918527
        },
        {
          "tokens": 16384,
          "tokens_per_second": 5377.969026899174
        }
      ],
      "decode": {
        "generated_tokens": 128,
        "tokens_per_second": 25.420647919679364
      }
    }
  }
}