{
  "schema": "computearena-public-benchmark/1",
  "submission_id": "98ac889f-306b-4097-adc8-f9b79033ac18",
  "run_id": "bc1698af38294b6e29df418b5ca4dbff",
  "created_at_unix_ms": 1789863423931,
  "submitted_by": "prabod",
  "model": {
    "canonical_id": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
    "identity_resolution": "artifact_verified",
    "identity_verification": "verified",
    "verified_at": "2026-09-11T12:18:35.929Z",
    "name": "NVIDIA-Nemotron-3-Nano-30B-A3B-Q4",
    "architecture": "nemotron_h_moe",
    "quantization": "base_q4",
    "artifact": {
      "provider": "huggingface",
      "repo_id": "basecompute/NVIDIA-Nemotron-3-Nano-30B-A3B",
      "revision": "93f8e8648a960d4c5d33663ea282ef7a3981b7fc",
      "path": "NVIDIA-Nemotron-3-Nano-30B-A3B-Q4.base",
      "sha256": "1b13d5d33fdc7312ca6ac1fd23bc41c74c2fddc3219e038bf6060a7e6dae28fb",
      "format": "base",
      "quantization_namespace": "basert",
      "quantization_scheme": "base_q4"
    }
  },
  "runtime": {
    "name": "basert",
    "version": "0.2.6"
  },
  "benchmark": {
    "mode": "text",
    "chip": "Apple M1 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": 608.4148263357321
        },
        {
          "tokens": 256,
          "tokens_per_second": 728.4650407710552
        },
        {
          "tokens": 512,
          "tokens_per_second": 870.1116474174138
        },
        {
          "tokens": 1024,
          "tokens_per_second": 943.7696698317582
        },
        {
          "tokens": 2048,
          "tokens_per_second": 977.6093559243849
        },
        {
          "tokens": 4096,
          "tokens_per_second": 970.848050617445
        },
        {
          "tokens": 8192,
          "tokens_per_second": 946.6246961798023
        },
        {
          "tokens": 16384,
          "tokens_per_second": 894.0736245995905
        }
      ],
      "decode": {
        "generated_tokens": 128,
        "tokens_per_second": 110.33541131490597
      }
    }
  }
}