{
  "caveats": [
    "Finite lab cohort; configured 262144 context differs from measured inputs. Aggregate uses first-output to final-completion window. Whole-host memory sampled during speed only; missing counters remain unassessed. Broader coding quality and delivery acceptance unassessed."
  ],
  "engine": {
    "name": "NVIDIA NGC vLLM",
    "recipe": "plain",
    "version": "0.29.0+5013de39.dev"
  },
  "format": "olir/0.1",
  "history": {
    "corrections": [],
    "supersedes": []
  },
  "id": "5366b32d63594303bd6c627d",
  "measured_at": "2026-10-07T05:38:56.618035+00:00",
  "metrics": [
    {
      "basis": null,
      "n": null,
      "name": "wall_seconds",
      "scope": "machine_total",
      "source_field": "wall_seconds",
      "statistic": "single",
      "unit": "s",
      "value": 85.29438710212708
    },
    {
      "basis": null,
      "n": 8,
      "name": "ttft_median_s",
      "scope": "per_person",
      "source_field": "ttft_median_s",
      "statistic": "median",
      "unit": "s",
      "value": 54.295966231497005
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_total",
      "scope": "machine_total",
      "source_field": "needle_total",
      "statistic": "single",
      "unit": "count",
      "value": 24
    },
    {
      "basis": null,
      "n": 8,
      "name": "ttft_max_s",
      "scope": "per_person",
      "source_field": "ttft_max_s",
      "statistic": "max",
      "unit": "s",
      "value": 63.35083819000283
    },
    {
      "basis": null,
      "n": null,
      "name": "requests_ok",
      "scope": "machine_total",
      "source_field": "requests_ok",
      "statistic": "single",
      "unit": "count",
      "value": 8
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_correct",
      "scope": "machine_total",
      "source_field": "needle_correct",
      "statistic": "single",
      "unit": "count",
      "value": 18
    },
    {
      "basis": null,
      "n": null,
      "name": "preemptions",
      "scope": "machine_total",
      "source_field": "preemptions",
      "statistic": "single",
      "unit": "count",
      "value": 0
    },
    {
      "basis": null,
      "n": null,
      "name": "decode_overlap",
      "scope": "machine_total",
      "source_field": "decode_overlap",
      "statistic": "single",
      "unit": "count",
      "value": 8
    },
    {
      "basis": null,
      "n": null,
      "name": "aggregate_concurrency",
      "scope": "machine_total",
      "source_field": "aggregate_concurrency",
      "statistic": "single",
      "unit": "count",
      "value": 8
    },
    {
      "basis": null,
      "n": null,
      "name": "requests_total",
      "scope": "machine_total",
      "source_field": "requests_total",
      "statistic": "single",
      "unit": "count",
      "value": 8
    },
    {
      "basis": null,
      "n": null,
      "name": "correctness_probes_passed",
      "scope": "machine_total",
      "source_field": "correctness_probes_passed",
      "statistic": "single",
      "unit": "count",
      "value": 6
    },
    {
      "basis": null,
      "n": null,
      "name": "aggregate_decode_tps",
      "scope": "machine_total",
      "source_field": "aggregate_tps",
      "statistic": "single",
      "unit": "tok/s",
      "value": 125.33113262551221
    },
    {
      "basis": null,
      "n": null,
      "name": "aggregate_e2e_tps",
      "scope": "machine_total",
      "source_field": "aggregate_end_to_end_tps",
      "statistic": "single",
      "unit": "tok/s",
      "value": 48.02191725811526
    },
    {
      "basis": null,
      "n": null,
      "name": "correctness_probes_total",
      "scope": "machine_total",
      "source_field": "correctness_probes_total",
      "statistic": "single",
      "unit": "count",
      "value": 8
    },
    {
      "basis": null,
      "n": 8,
      "name": "decode_tps",
      "scope": "per_person",
      "source_field": "per_request_decode_tps",
      "statistic": "mean",
      "unit": "tok/s",
      "value": 18.18877871935077
    },
    {
      "basis": null,
      "n": null,
      "name": "forced_output_tokens",
      "scope": "machine_total",
      "source_field": "forced_output_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 512
    },
    {
      "basis": null,
      "n": null,
      "name": "peak_host_unavailable_gib",
      "scope": "machine_total",
      "source_field": "peak_host_unavailable_gib",
      "statistic": "max",
      "unit": "GiB",
      "value": 94.6588134765625
    },
    {
      "basis": null,
      "n": null,
      "name": "ttft_s",
      "scope": "per_person",
      "source_field": "team_load_requests[0].ttft_s",
      "statistic": "single",
      "unit": "s",
      "value": 54.29679778503487
    },
    {
      "basis": null,
      "n": null,
      "name": "total_s",
      "scope": "per_person",
      "source_field": "team_load_requests[0].total_s",
      "statistic": "single",
      "unit": "s",
      "value": 84.91481842304347
    },
    {
      "basis": null,
      "n": null,
      "name": "decode_tps",
      "scope": "per_person",
      "source_field": "team_load_requests[0].decode_tps",
      "statistic": "single",
      "unit": "tok/s",
      "value": 16.68951778566818
    },
    {
      "basis": null,
      "n": null,
      "name": "user_number",
      "scope": "per_person",
      "source_field": "team_load_requests[0].user_number",
      "statistic": "single",
      "unit": "count",
      "value": 0
    },
    {
      "basis": null,
      "n": null,
      "name": "prompt_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[0].prompt_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 40096
    },
    {
      "basis": null,
      "n": null,
      "name": "output_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[0].output_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 512
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_correct",
      "scope": "per_person",
      "source_field": "team_load_requests[0].needle_correct",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_total",
      "scope": "per_person",
      "source_field": "team_load_requests[0].needle_total",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "ttft_s",
      "scope": "per_person",
      "source_field": "team_load_requests[1].ttft_s",
      "statistic": "single",
      "unit": "s",
      "value": 54.295235718018375
    },
    {
      "basis": null,
      "n": null,
      "name": "total_s",
      "scope": "per_person",
      "source_field": "team_load_requests[1].total_s",
      "statistic": "single",
      "unit": "s",
      "value": 84.91270488203736
    },
    {
      "basis": null,
      "n": null,
      "name": "decode_tps",
      "scope": "per_person",
      "source_field": "team_load_requests[1].decode_tps",
      "statistic": "single",
      "unit": "tok/s",
      "value": 16.68981839297536
    },
    {
      "basis": null,
      "n": null,
      "name": "user_number",
      "scope": "per_person",
      "source_field": "team_load_requests[1].user_number",
      "statistic": "single",
      "unit": "count",
      "value": 1
    },
    {
      "basis": null,
      "n": null,
      "name": "prompt_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[1].prompt_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 40103
    },
    {
      "basis": null,
      "n": null,
      "name": "output_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[1].output_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 512
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_correct",
      "scope": "per_person",
      "source_field": "team_load_requests[1].needle_correct",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_total",
      "scope": "per_person",
      "source_field": "team_load_requests[1].needle_total",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "ttft_s",
      "scope": "per_person",
      "source_field": "team_load_requests[2].ttft_s",
      "statistic": "single",
      "unit": "s",
      "value": 63.26338509900961
    },
    {
      "basis": null,
      "n": null,
      "name": "total_s",
      "scope": "per_person",
      "source_field": "team_load_requests[2].total_s",
      "statistic": "single",
      "unit": "s",
      "value": 85.28199579700595
    },
    {
      "basis": null,
      "n": null,
      "name": "decode_tps",
      "scope": "per_person",
      "source_field": "team_load_requests[2].decode_tps",
      "statistic": "single",
      "unit": "tok/s",
      "value": 23.2076404369373
    },
    {
      "basis": null,
      "n": null,
      "name": "user_number",
      "scope": "per_person",
      "source_field": "team_load_requests[2].user_number",
      "statistic": "single",
      "unit": "count",
      "value": 2
    },
    {
      "basis": null,
      "n": null,
      "name": "prompt_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[2].prompt_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 40102
    },
    {
      "basis": null,
      "n": null,
      "name": "output_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[2].output_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 512
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_correct",
      "scope": "per_person",
      "source_field": "team_load_requests[2].needle_correct",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_total",
      "scope": "per_person",
      "source_field": "team_load_requests[2].needle_total",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "ttft_s",
      "scope": "per_person",
      "source_field": "team_load_requests[3].ttft_s",
      "statistic": "single",
      "unit": "s",
      "value": 54.29382729099598
    },
    {
      "basis": null,
      "n": null,
      "name": "total_s",
      "scope": "per_person",
      "source_field": "team_load_requests[3].total_s",
      "statistic": "single",
      "unit": "s",
      "value": 84.9107472699834
    },
    {
      "basis": null,
      "n": null,
      "name": "decode_tps",
      "scope": "per_person",
      "source_field": "team_load_requests[3].decode_tps",
      "statistic": "single",
      "unit": "tok/s",
      "value": 16.690117763338126
    },
    {
      "basis": null,
      "n": null,
      "name": "user_number",
      "scope": "per_person",
      "source_field": "team_load_requests[3].user_number",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "prompt_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[3].prompt_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 40105
    },
    {
      "basis": null,
      "n": null,
      "name": "output_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[3].output_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 512
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_correct",
      "scope": "per_person",
      "source_field": "team_load_requests[3].needle_correct",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_total",
      "scope": "per_person",
      "source_field": "team_load_requests[3].needle_total",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "ttft_s",
      "scope": "per_person",
      "source_field": "team_load_requests[4].ttft_s",
      "statistic": "single",
      "unit": "s",
      "value": 63.35083819000283
    },
    {
      "basis": null,
      "n": null,
      "name": "total_s",
      "scope": "per_person",
      "source_field": "team_load_requests[4].total_s",
      "statistic": "single",
      "unit": "s",
      "value": 85.29077736200998
    },
    {
      "basis": null,
      "n": null,
      "name": "decode_tps",
      "scope": "per_person",
      "source_field": "team_load_requests[4].decode_tps",
      "statistic": "single",
      "unit": "tok/s",
      "value": 23.290857645219802
    },
    {
      "basis": null,
      "n": null,
      "name": "user_number",
      "scope": "per_person",
      "source_field": "team_load_requests[4].user_number",
      "statistic": "single",
      "unit": "count",
      "value": 4
    },
    {
      "basis": null,
      "n": null,
      "name": "prompt_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[4].prompt_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 40097
    },
    {
      "basis": null,
      "n": null,
      "name": "output_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[4].output_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 512
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_correct",
      "scope": "per_person",
      "source_field": "team_load_requests[4].needle_correct",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_total",
      "scope": "per_person",
      "source_field": "team_load_requests[4].needle_total",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "ttft_s",
      "scope": "per_person",
      "source_field": "team_load_requests[5].ttft_s",
      "statistic": "single",
      "unit": "s",
      "value": 52.676520195032936
    },
    {
      "basis": null,
      "n": null,
      "name": "total_s",
      "scope": "per_person",
      "source_field": "team_load_requests[5].total_s",
      "statistic": "single",
      "unit": "s",
      "value": 84.83710313600022
    },
    {
      "basis": null,
      "n": null,
      "name": "decode_tps",
      "scope": "per_person",
      "source_field": "team_load_requests[5].decode_tps",
      "statistic": "single",
      "unit": "tok/s",
      "value": 15.889015473941242
    },
    {
      "basis": null,
      "n": null,
      "name": "user_number",
      "scope": "per_person",
      "source_field": "team_load_requests[5].user_number",
      "statistic": "single",
      "unit": "count",
      "value": 5
    },
    {
      "basis": null,
      "n": null,
      "name": "prompt_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[5].prompt_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 40104
    },
    {
      "basis": null,
      "n": null,
      "name": "output_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[5].output_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 512
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_correct",
      "scope": "per_person",
      "source_field": "team_load_requests[5].needle_correct",
      "statistic": "single",
      "unit": "count",
      "value": 0
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_total",
      "scope": "per_person",
      "source_field": "team_load_requests[5].needle_total",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "ttft_s",
      "scope": "per_person",
      "source_field": "team_load_requests[6].ttft_s",
      "statistic": "single",
      "unit": "s",
      "value": 53.644833061960526
    },
    {
      "basis": null,
      "n": null,
      "name": "total_s",
      "scope": "per_person",
      "source_field": "team_load_requests[6].total_s",
      "statistic": "single",
      "unit": "s",
      "value": 84.87352550000651
    },
    {
      "basis": null,
      "n": null,
      "name": "decode_tps",
      "scope": "per_person",
      "source_field": "team_load_requests[6].decode_tps",
      "statistic": "single",
      "unit": "tok/s",
      "value": 16.363157087468945
    },
    {
      "basis": null,
      "n": null,
      "name": "user_number",
      "scope": "per_person",
      "source_field": "team_load_requests[6].user_number",
      "statistic": "single",
      "unit": "count",
      "value": 6
    },
    {
      "basis": null,
      "n": null,
      "name": "prompt_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[6].prompt_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 40105
    },
    {
      "basis": null,
      "n": null,
      "name": "output_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[6].output_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 512
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_correct",
      "scope": "per_person",
      "source_field": "team_load_requests[6].needle_correct",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_total",
      "scope": "per_person",
      "source_field": "team_load_requests[6].needle_total",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": null,
      "name": "ttft_s",
      "scope": "per_person",
      "source_field": "team_load_requests[7].ttft_s",
      "statistic": "single",
      "unit": "s",
      "value": 54.296696744975634
    },
    {
      "basis": null,
      "n": null,
      "name": "total_s",
      "scope": "per_person",
      "source_field": "team_load_requests[7].total_s",
      "statistic": "single",
      "unit": "s",
      "value": 84.91363982699113
    },
    {
      "basis": null,
      "n": null,
      "name": "decode_tps",
      "scope": "per_person",
      "source_field": "team_load_requests[7].decode_tps",
      "statistic": "single",
      "unit": "tok/s",
      "value": 16.69010516925719
    },
    {
      "basis": null,
      "n": null,
      "name": "user_number",
      "scope": "per_person",
      "source_field": "team_load_requests[7].user_number",
      "statistic": "single",
      "unit": "count",
      "value": 7
    },
    {
      "basis": null,
      "n": null,
      "name": "prompt_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[7].prompt_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 40097
    },
    {
      "basis": null,
      "n": null,
      "name": "output_tokens",
      "scope": "per_person",
      "source_field": "team_load_requests[7].output_tokens",
      "statistic": "single",
      "unit": "tokens",
      "value": 512
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_correct",
      "scope": "per_person",
      "source_field": "team_load_requests[7].needle_correct",
      "statistic": "single",
      "unit": "count",
      "value": 0
    },
    {
      "basis": null,
      "n": null,
      "name": "needle_total",
      "scope": "per_person",
      "source_field": "team_load_requests[7].needle_total",
      "statistic": "single",
      "unit": "count",
      "value": 3
    },
    {
      "basis": null,
      "n": 8,
      "name": "decode_tps",
      "scope": "per_person",
      "source_field": "per_person_decode_median_tps",
      "statistic": "median",
      "unit": "tok/s",
      "value": 16.689961781116274
    }
  ],
  "model": {
    "artifact_sha256": [
      "648fe83171485f4be03ea58d86089e088c94c2a3dc88922233f9002f30d8bc8b",
      "8303cef7d375021d69dcafc60d3fad5f8186995b18b0db5e2a609aadeb8a4f51",
      "cb5c5978721a1e87406ae68f1cd5a855326e40ff8bc13832644c392115336987"
    ],
    "publisher_repo": "ornith-ai/Ornith-1.5-35B-A3B",
    "publisher_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
    "variant": "NVFP4 variant",
    "weights_dtype": "nvfp4",
    "weights_format": "safetensors"
  },
  "outcome": {
    "gates": [
      {
        "met": true,
        "name": "requests",
        "observed": "8/8"
      },
      {
        "met": false,
        "name": "hidden-facts",
        "observed": "18/24"
      },
      {
        "met": false,
        "name": "correctness-probes",
        "observed": "6/8"
      }
    ],
    "reasons": [
      "18/24 Hidden facts found.",
      "6/8 probes passed."
    ],
    "status": "fail"
  },
  "publication_group": "pre-standard",
  "published": null,
  "publisher": {
    "name": "Skilak Keep"
  },
  "quality": [
    {
      "passed": 18,
      "skipped": null,
      "suite": "hidden-facts",
      "total": 24,
      "version": "1"
    },
    {
      "passed": 6,
      "skipped": null,
      "suite": "correctness-probes",
      "total": 8,
      "version": "1"
    }
  ],
  "record_date": "2026-10-07",
  "reproduce": {
    "driver": null,
    "fixture_sha256": "e2097959b1ca0142f2b93edc3ec411e474c0a6622800dd247af7a71bd9c39e83",
    "harness": {
      "name": "keep-team-load-v1",
      "version": "1"
    },
    "harness_sha256": null,
    "os": null,
    "workload_name": null,
    "workload_version": null
  },
  "requests": [
    {
      "cached_prompt_tokens": null,
      "checks_passed": true,
      "completed": true,
      "decode_tps": 16.68951778566818,
      "effective_prefill_tps": null,
      "needle_correct": 3,
      "needle_total": 3,
      "output_tokens": 512,
      "prompt_tokens": 40096,
      "total_s": 84.91481842304347,
      "ttft_s": 54.29679778503487,
      "user_number": 0
    },
    {
      "cached_prompt_tokens": null,
      "checks_passed": true,
      "completed": true,
      "decode_tps": 16.68981839297536,
      "effective_prefill_tps": null,
      "needle_correct": 3,
      "needle_total": 3,
      "output_tokens": 512,
      "prompt_tokens": 40103,
      "total_s": 84.91270488203736,
      "ttft_s": 54.295235718018375,
      "user_number": 1
    },
    {
      "cached_prompt_tokens": null,
      "checks_passed": true,
      "completed": true,
      "decode_tps": 23.2076404369373,
      "effective_prefill_tps": null,
      "needle_correct": 3,
      "needle_total": 3,
      "output_tokens": 512,
      "prompt_tokens": 40102,
      "total_s": 85.28199579700595,
      "ttft_s": 63.26338509900961,
      "user_number": 2
    },
    {
      "cached_prompt_tokens": null,
      "checks_passed": true,
      "completed": true,
      "decode_tps": 16.690117763338126,
      "effective_prefill_tps": null,
      "needle_correct": 3,
      "needle_total": 3,
      "output_tokens": 512,
      "prompt_tokens": 40105,
      "total_s": 84.9107472699834,
      "ttft_s": 54.29382729099598,
      "user_number": 3
    },
    {
      "cached_prompt_tokens": null,
      "checks_passed": true,
      "completed": true,
      "decode_tps": 23.290857645219802,
      "effective_prefill_tps": null,
      "needle_correct": 3,
      "needle_total": 3,
      "output_tokens": 512,
      "prompt_tokens": 40097,
      "total_s": 85.29077736200998,
      "ttft_s": 63.35083819000283,
      "user_number": 4
    },
    {
      "cached_prompt_tokens": null,
      "checks_passed": false,
      "completed": true,
      "decode_tps": 15.889015473941242,
      "effective_prefill_tps": null,
      "needle_correct": 0,
      "needle_total": 3,
      "output_tokens": 512,
      "prompt_tokens": 40104,
      "total_s": 84.83710313600022,
      "ttft_s": 52.676520195032936,
      "user_number": 5
    },
    {
      "cached_prompt_tokens": null,
      "checks_passed": true,
      "completed": true,
      "decode_tps": 16.363157087468945,
      "effective_prefill_tps": null,
      "needle_correct": 3,
      "needle_total": 3,
      "output_tokens": 512,
      "prompt_tokens": 40105,
      "total_s": 84.87352550000651,
      "ttft_s": 53.644833061960526,
      "user_number": 6
    },
    {
      "cached_prompt_tokens": null,
      "checks_passed": false,
      "completed": true,
      "decode_tps": 16.69010516925719,
      "effective_prefill_tps": null,
      "needle_correct": 0,
      "needle_total": 3,
      "output_tokens": 512,
      "prompt_tokens": 40097,
      "total_s": 84.91363982699113,
      "ttft_s": 54.296696744975634,
      "user_number": 7
    }
  ],
  "scenario": {
    "id": "team-load",
    "version": "1"
  },
  "settings": {
    "concurrency": 8,
    "configured_context_tokens": 262144,
    "context_tokens": 262144,
    "kv_dtype": "fp8",
    "max_output_tokens": 512,
    "speculative_decoding": "plain",
    "thinking": false
  },
  "setup_complete": false,
  "setup_id": "a167e4bafd3fd96d0b47cf9d6645473d",
  "standard": "pre-standard",
  "system": {
    "driver": null,
    "hardware": "NVIDIA GB10, 128 GB unified memory",
    "os": null
  },
  "workload": {
    "cache": null,
    "content_tokens": null,
    "document": null,
    "input_tokens": {
      "max": 40105,
      "median": 40102.5,
      "min": 40096
    },
    "output_tokens": 512,
    "repetitions": null,
    "users": 8,
    "warmup": null
  }
}
