{
  "schema": "computearena-public-benchmark/1",
  "submission_id": "5ed12878-4bcc-4680-9023-5f9f6322fefc",
  "run_id": "023f6bc3247ecdbb4055dbeab8c5614f",
  "created_at_unix_ms": 1789840057661,
  "submitted_by": "prabod",
  "model": {
    "canonical_id": "openai/gpt-oss-20b",
    "identity_resolution": "artifact_verified",
    "identity_verification": "verified",
    "verified_at": "2026-09-11T12:18:35.929Z",
    "name": "gpt-oss-20b-Q4",
    "architecture": "gpt_oss",
    "quantization": "base_q4",
    "artifact": {
      "provider": "huggingface",
      "repo_id": "basecompute/gpt-oss-20b",
      "revision": "6dc70937996ff2cf3bb552046f693018b43f7cc8",
      "path": "gpt-oss-20b-Q4.base",
      "sha256": "6b7ae00571be1b9d4e5204c34e9fe22c19ee69c5dd6e51d32900cf8d41091c7f",
      "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": 2387.037102543094
        },
        {
          "tokens": 256,
          "tokens_per_second": 3462.0925007402475
        },
        {
          "tokens": 512,
          "tokens_per_second": 4292.445037195247
        },
        {
          "tokens": 1024,
          "tokens_per_second": 4806.662022758644
        },
        {
          "tokens": 2048,
          "tokens_per_second": 4772.52453596724
        },
        {
          "tokens": 4096,
          "tokens_per_second": 4238.87318026782
        },
        {
          "tokens": 8192,
          "tokens_per_second": 3425.1422969450955
        },
        {
          "tokens": 16384,
          "tokens_per_second": 2467.130138310517
        }
      ],
      "decode": {
        "generated_tokens": 128,
        "tokens_per_second": 57.343320335060525
      }
    }
  }
}