{
  "schema": "computearena-public-benchmark/1",
  "submission_id": "50eb8c2d-e6d3-4e99-9f72-14d171c675ee",
  "run_id": "8b9a39f7eb17ea50794dcf8360b9311b",
  "created_at_unix_ms": 1789839203162,
  "submitted_by": "prabod",
  "model": {
    "canonical_id": "Qwen/Qwen3-4B-Instruct-2507",
    "identity_resolution": "artifact_verified",
    "identity_verification": "verified",
    "verified_at": "2026-09-11T12:18:35.929Z",
    "name": "Qwen3-4B-Instruct-2507-Q8",
    "architecture": "qwen",
    "quantization": "base_q8",
    "artifact": {
      "provider": "huggingface",
      "repo_id": "basecompute/Qwen3-4B-Instruct-2507",
      "revision": "a31561f6fd063baa676df238611fc7df3b473f35",
      "path": "Qwen3-4B-Instruct-2507-Q8.base",
      "sha256": "09f7ac220735baf95d4effcf85417c8c8eadf69a42d21e48edcf33924ea55a40",
      "format": "base",
      "quantization_namespace": "basert",
      "quantization_scheme": "base_q8"
    }
  },
  "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": 4731.620810400309
        },
        {
          "tokens": 256,
          "tokens_per_second": 8035.875176471726
        },
        {
          "tokens": 512,
          "tokens_per_second": 10969.97320134297
        },
        {
          "tokens": 1024,
          "tokens_per_second": 12760.131442053702
        },
        {
          "tokens": 2048,
          "tokens_per_second": 12277.489813184611
        },
        {
          "tokens": 4096,
          "tokens_per_second": 10924.630705463773
        },
        {
          "tokens": 8192,
          "tokens_per_second": 8975.369744285616
        },
        {
          "tokens": 16384,
          "tokens_per_second": 6652.465308532027
        }
      ],
      "decode": {
        "generated_tokens": 128,
        "tokens_per_second": 54.786424612263495
      }
    }
  }
}