{
  "schema": "computearena-public-benchmark/1",
  "submission_id": "85abfc06-3563-44a3-b16b-844c6ac63b30",
  "run_id": "530b9b93fa499ba27ec8cbc638cc0879",
  "created_at_unix_ms": 1789838894690,
  "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-Q4",
    "architecture": "qwen",
    "quantization": "base_q4",
    "artifact": {
      "provider": "huggingface",
      "repo_id": "basecompute/Qwen3-4B-Instruct-2507",
      "revision": "a31561f6fd063baa676df238611fc7df3b473f35",
      "path": "Qwen3-4B-Instruct-2507-Q4.base",
      "sha256": "4523c9d3f10dc77bd6e82a78dee495f0a5241f0a4a1eb3902216ec0df6bf94b9",
      "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": 4748.28453073272
        },
        {
          "tokens": 256,
          "tokens_per_second": 7837.6731351137605
        },
        {
          "tokens": 512,
          "tokens_per_second": 11246.34559153756
        },
        {
          "tokens": 1024,
          "tokens_per_second": 12611.930614910903
        },
        {
          "tokens": 2048,
          "tokens_per_second": 12112.057381153589
        },
        {
          "tokens": 4096,
          "tokens_per_second": 10837.431807964176
        },
        {
          "tokens": 8192,
          "tokens_per_second": 8956.640545316846
        },
        {
          "tokens": 16384,
          "tokens_per_second": 6655.956090402656
        }
      ],
      "decode": {
        "generated_tokens": 128,
        "tokens_per_second": 73.69889163145004
      }
    }
  }
}