{
  "schema": "computearena-public-benchmark/1",
  "submission_id": "d0c80da0-a6a4-492c-8c2d-936e53d214fc",
  "run_id": "92e4d7ffbbbc4d0789e63cdab1056641",
  "created_at_unix_ms": 1789839754163,
  "submitted_by": "prabod",
  "model": {
    "canonical_id": "mistralai/Mistral-7B-Instruct-v0.3",
    "identity_resolution": "artifact_verified",
    "identity_verification": "verified",
    "verified_at": "2026-09-11T12:18:35.929Z",
    "name": "Mistral-7B-Instruct-v0.3-Q8",
    "architecture": "llama",
    "quantization": "base_q4",
    "artifact": {
      "provider": "huggingface",
      "repo_id": "basecompute/Mistral-7B-Instruct-v0.3",
      "revision": "3805d321f571cdbb8399bc13d9baa7ef068582b9",
      "path": "Mistral-7B-Instruct-v0.3-Q8.base",
      "sha256": "34a0f7e8b6fda223a39b6b632009ec19503db794bfa48ab7ec531dad9a37ef2f",
      "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": 2963.722642701024
        },
        {
          "tokens": 256,
          "tokens_per_second": 5134.851427862767
        },
        {
          "tokens": 512,
          "tokens_per_second": 7510.185624207846
        },
        {
          "tokens": 1024,
          "tokens_per_second": 8171.186211954379
        },
        {
          "tokens": 2048,
          "tokens_per_second": 7919.5193677241405
        },
        {
          "tokens": 4096,
          "tokens_per_second": 7308.685385383314
        },
        {
          "tokens": 8192,
          "tokens_per_second": 6471.788364019674
        },
        {
          "tokens": 16384,
          "tokens_per_second": 5253.00325393114
        }
      ],
      "decode": {
        "generated_tokens": 128,
        "tokens_per_second": 26.977133067999873
      }
    }
  }
}