{
  "generated_at_utc": "2026-10-22T03:32:09.604144+00:00",
  "model": "nvidia/Qwen3.8-27B-NVFP4",
  "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
  "hardware": "One full96GB RTX PRO6000 Blackwell Server Edition",
  "candidate_gpu_count": 1,
  "production_modified": false,
  "model_variants": [
    {
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "revision": "482ca0f3832238542f8f5295dde86b5f22711d80"
    },
    {
      "model": "unsloth/Qwen3.8-27B-GGUF",
      "revision": "4ca720788d1e01f1bff70c033e0d0028fd02e502"
    }
  ],
  "study": "qwen38-nvfp4-2026-10-21",
  "status": "experimentation-ended-no-production-promotion",
  "source_measurements_sha256": "1fed2bb8c734f912c002bfba896d002e9a9d97e0e7f8c7e2ce3ea95951c2f890",
  "agentx": [
    {
      "config_id": "baseline-mtp3",
      "series": "direct-origin-agentx64k",
      "concurrency": 1,
      "requests": 79,
      "raw_span_output_tok_s": 52.867201905326446,
      "decode_tok_s_user": 128.89305145038998,
      "ttft_p95_ms": 2310.3641156999943,
      "request_latency_p95_ms": 19350.955394399996,
      "zero_request_errors": true,
      "native_aggregate_comparable": false,
      "raw_span_delta_percent": 0.0
    },
    {
      "config_id": "baseline-mtp3",
      "series": "candidate-origin-agentx64k",
      "concurrency": 1,
      "requests": 79,
      "raw_span_output_tok_s": 52.50381826383164,
      "decode_tok_s_user": 123.24622944245696,
      "ttft_p95_ms": 1937.746493599994,
      "request_latency_p95_ms": 20931.333879299993,
      "zero_request_errors": true,
      "native_aggregate_comparable": true,
      "raw_span_delta_percent": 0.0
    },
    {
      "config_id": "baseline-mtp3",
      "series": "candidate-origin-agentx64k",
      "concurrency": 8,
      "requests": 206,
      "raw_span_output_tok_s": 93.0629275206239,
      "decode_tok_s_user": 113.60995323775552,
      "ttft_p95_ms": 3217.82167725,
      "request_latency_p95_ms": 19328.505778,
      "zero_request_errors": true,
      "native_aggregate_comparable": false,
      "raw_span_delta_percent": 0.0
    },
    {
      "config_id": "baseline-mtp3",
      "series": "candidate-origin-agentx64k",
      "concurrency": 16,
      "requests": 283,
      "raw_span_output_tok_s": 129.0335659301832,
      "decode_tok_s_user": 101.38850926059082,
      "ttft_p95_ms": 4997.4891807999875,
      "request_latency_p95_ms": 22013.098089299983,
      "zero_request_errors": true,
      "native_aggregate_comparable": false,
      "raw_span_delta_percent": 0.0
    },
    {
      "config_id": "baseline-mtp3",
      "series": "candidate-loopback-agentx64k",
      "concurrency": 32,
      "requests": 376,
      "raw_span_output_tok_s": 171.18653912817285,
      "decode_tok_s_user": 52.74304726251263,
      "ttft_p95_ms": 67573.2816505,
      "request_latency_p95_ms": 103120.01512975,
      "zero_request_errors": true,
      "native_aggregate_comparable": false,
      "raw_span_delta_percent": 0.0
    },
    {
      "config_id": "eagle4-depth6-prefix-full",
      "series": "candidate-loopback-agentx64k",
      "concurrency": 32,
      "requests": 253,
      "raw_span_output_tok_s": 112.89759955125312,
      "decode_tok_s_user": 34.80104568685519,
      "ttft_p95_ms": 110381.35710579998,
      "request_latency_p95_ms": 174217.40911159993,
      "zero_request_errors": true,
      "native_aggregate_comparable": false,
      "raw_span_delta_percent": -34.04995502203414
    },
    {
      "config_id": "maxseq16",
      "series": "candidate-loopback-agentx64k",
      "concurrency": 32,
      "requests": 360,
      "raw_span_output_tok_s": 160.84477167283973,
      "decode_tok_s_user": 46.515117523334425,
      "ttft_p95_ms": 57795.442230550056,
      "request_latency_p95_ms": 128162.48566235005,
      "zero_request_errors": true,
      "native_aggregate_comparable": false,
      "raw_span_delta_percent": -6.041227019368566
    },
    {
      "config_id": "sglang-target-api-full",
      "series": "candidate-loopback-agentx64k",
      "concurrency": 32,
      "requests": 229,
      "raw_span_output_tok_s": 93.25235126668758,
      "decode_tok_s_user": 21.973416307205575,
      "ttft_p95_ms": 138104.305268,
      "request_latency_p95_ms": 214085.14386239997,
      "zero_request_errors": true,
      "native_aggregate_comparable": false,
      "raw_span_delta_percent": -45.5258855388936
    },
    {
      "config_id": "sglang-dflash2-api-full",
      "series": "candidate-loopback-agentx64k",
      "concurrency": 32,
      "requests": 240,
      "raw_span_output_tok_s": 104.59005495957075,
      "decode_tok_s_user": 35.152222044635316,
      "ttft_p95_ms": 109289.81035564998,
      "request_latency_p95_ms": 167083.65328615,
      "zero_request_errors": true,
      "native_aggregate_comparable": false,
      "raw_span_delta_percent": -38.90287431930567
    },
    {
      "config_id": "openjet-unsloth-udq4xl-mtp4-retry-20261022",
      "series": "candidate-loopback-agentx64k-gguf-openjet",
      "concurrency": 32,
      "requests": 84,
      "raw_span_output_tok_s": 36.51403337484811,
      "decode_tok_s_user": 147.32099461504862,
      "ttft_p95_ms": 483911.53780039994,
      "request_latency_p95_ms": 486075.14030529995,
      "zero_request_errors": true,
      "native_aggregate_comparable": false
    }
  ],
  "synthetic": [
    {
      "config_id": "baseline-mtp3",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 1,
      "valid_runs": 1,
      "median_raw_span_tok_s": 146.08578906025147,
      "observed_range_tok_s": [
        146.08578906025147,
        146.08578906025147
      ],
      "median_run_ttft_p95_ms": 413.52696554999994,
      "median_run_e2e_p95_ms": 8449.5410375,
      "median_run_decode_tok_s_per_user": 154.39563159794548,
      "delta_percent": 0.9994571625319582
    },
    {
      "config_id": "baseline-mtp3",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 8,
      "valid_runs": 1,
      "median_raw_span_tok_s": 1008.5343429828421,
      "observed_range_tok_s": [
        1008.5343429828421,
        1008.5343429828421
      ],
      "median_run_ttft_p95_ms": 550.5722340499999,
      "median_run_e2e_p95_ms": 10264.022802049998,
      "median_run_decode_tok_s_per_user": 135.9464884070363,
      "delta_percent": 0.7364010872153548
    },
    {
      "config_id": "baseline-mtp3",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 16,
      "valid_runs": 1,
      "median_raw_span_tok_s": 1053.112799776047,
      "observed_range_tok_s": [
        1053.112799776047,
        1053.112799776047
      ],
      "median_run_ttft_p95_ms": 9041.13133075,
      "median_run_e2e_p95_ms": 17562.7202955,
      "median_run_decode_tok_s_per_user": 138.49502053799938,
      "delta_percent": 5.557458602319243
    },
    {
      "config_id": "baseline-mtp3",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 32,
      "valid_runs": 1,
      "median_raw_span_tok_s": 1016.5140638801491,
      "observed_range_tok_s": [
        1016.5140638801491,
        1016.5140638801491
      ],
      "median_run_ttft_p95_ms": 25671.280315999997,
      "median_run_e2e_p95_ms": 33977.4459964,
      "median_run_decode_tok_s_per_user": 135.05430339783132,
      "delta_percent": 1.4573151056927136
    },
    {
      "config_id": "baseline-mtp3-recheck",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 1,
      "valid_runs": 1,
      "median_raw_span_tok_s": 144.64017249633824,
      "observed_range_tok_s": [
        144.64017249633824,
        144.64017249633824
      ],
      "median_run_ttft_p95_ms": 480.67925239999994,
      "median_run_e2e_p95_ms": 8390.572457799999,
      "median_run_decode_tok_s_per_user": 155.02360336671197,
      "delta_percent": 0.0
    },
    {
      "config_id": "baseline-mtp3-recheck",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 8,
      "valid_runs": 1,
      "median_raw_span_tok_s": 1001.1617767738946,
      "observed_range_tok_s": [
        1001.1617767738946,
        1001.1617767738946
      ],
      "median_run_ttft_p95_ms": 639.5183321499984,
      "median_run_e2e_p95_ms": 10046.61772045,
      "median_run_decode_tok_s_per_user": 134.65135600454946,
      "delta_percent": 0.0
    },
    {
      "config_id": "baseline-mtp3-recheck",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 16,
      "valid_runs": 1,
      "median_raw_span_tok_s": 997.6678234965659,
      "observed_range_tok_s": [
        997.6678234965659,
        997.6678234965659
      ],
      "median_run_ttft_p95_ms": 9523.6414715,
      "median_run_e2e_p95_ms": 18243.639959,
      "median_run_decode_tok_s_per_user": 133.25656459729709,
      "delta_percent": 0.0
    },
    {
      "config_id": "baseline-mtp3-recheck",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 32,
      "valid_runs": 1,
      "median_raw_span_tok_s": 1001.9130338913463,
      "observed_range_tok_s": [
        1001.9130338913463,
        1001.9130338913463
      ],
      "median_run_ttft_p95_ms": 26430.65664775,
      "median_run_e2e_p95_ms": 34923.48363025,
      "median_run_decode_tok_s_per_user": 133.40232700440671,
      "delta_percent": 0.0
    },
    {
      "config_id": "eagle4-depth6-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 1,
      "valid_runs": 1,
      "median_raw_span_tok_s": 220.07633354688315,
      "observed_range_tok_s": [
        220.07633354688315,
        220.07633354688315
      ],
      "median_run_ttft_p95_ms": 390.28978474999997,
      "median_run_e2e_p95_ms": 6809.110488249999,
      "median_run_decode_tok_s_per_user": 244.38070486065757,
      "delta_percent": 52.15436330626242
    },
    {
      "config_id": "eagle4-depth6-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 8,
      "valid_runs": 1,
      "median_raw_span_tok_s": 1072.0043802893056,
      "observed_range_tok_s": [
        1072.0043802893056,
        1072.0043802893056
      ],
      "median_run_ttft_p95_ms": 604.1504941999999,
      "median_run_e2e_p95_ms": 10914.634891799999,
      "median_run_decode_tok_s_per_user": 154.13822838871877,
      "delta_percent": 7.07603957311389
    },
    {
      "config_id": "eagle4-depth6-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 16,
      "valid_runs": 1,
      "median_raw_span_tok_s": 1124.0116719207674,
      "observed_range_tok_s": [
        1124.0116719207674,
        1124.0116719207674
      ],
      "median_run_ttft_p95_ms": 8985.67726925,
      "median_run_e2e_p95_ms": 18122.10905275,
      "median_run_decode_tok_s_per_user": 154.73423088427361,
      "delta_percent": 12.663919337540541
    },
    {
      "config_id": "eagle4-depth6-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 32,
      "valid_runs": 1,
      "median_raw_span_tok_s": 1104.057753607387,
      "observed_range_tok_s": [
        1104.057753607387,
        1104.057753607387
      ],
      "median_run_ttft_p95_ms": 23768.621023149997,
      "median_run_e2e_p95_ms": 32279.00018945,
      "median_run_decode_tok_s_per_user": 153.81612010734466,
      "delta_percent": 10.194968651053383
    },
    {
      "config_id": "eagle4-depth6-prefix-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 1,
      "valid_runs": 1,
      "median_raw_span_tok_s": 199.55151459667786,
      "observed_range_tok_s": [
        199.55151459667786,
        199.55151459667786
      ],
      "median_run_ttft_p95_ms": 450.46808589999995,
      "median_run_e2e_p95_ms": 7074.5884648500005,
      "median_run_decode_tok_s_per_user": 226.91362278714965,
      "delta_percent": 37.96410164107746
    },
    {
      "config_id": "eagle4-depth6-prefix-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 8,
      "valid_runs": 1,
      "median_raw_span_tok_s": 1042.0949086468947,
      "observed_range_tok_s": [
        1042.0949086468947,
        1042.0949086468947
      ],
      "median_run_ttft_p95_ms": 728.5260056499999,
      "median_run_e2e_p95_ms": 10472.709891599996,
      "median_run_decode_tok_s_per_user": 150.9418677069148,
      "delta_percent": 4.0885631895478
    },
    {
      "config_id": "eagle4-depth6-reasoning-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 1,
      "valid_runs": 1,
      "median_raw_span_tok_s": 220.79895142755774,
      "observed_range_tok_s": [
        220.79895142755774,
        220.79895142755774
      ],
      "median_run_ttft_p95_ms": 355.470232,
      "median_run_e2e_p95_ms": 6743.1558595,
      "median_run_decode_tok_s_per_user": 244.86494545716033,
      "delta_percent": 52.65396024963089
    },
    {
      "config_id": "eagle4-depth6-structured-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 1,
      "valid_runs": 1,
      "median_raw_span_tok_s": 218.69901333591022,
      "observed_range_tok_s": [
        218.69901333591022,
        218.69901333591022
      ],
      "median_run_ttft_p95_ms": 516.983295,
      "median_run_e2e_p95_ms": 6731.2583589999995,
      "median_run_decode_tok_s_per_user": 248.0516902704227,
      "delta_percent": 51.2021242517855
    },
    {
      "config_id": "eagle4-isolated-frontend-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 1,
      "valid_runs": 1,
      "median_raw_span_tok_s": 147.52204851845474,
      "observed_range_tok_s": [
        147.52204851845474,
        147.52204851845474
      ],
      "median_run_ttft_p95_ms": 397.18297405,
      "median_run_e2e_p95_ms": 7498.855746849999,
      "median_run_decode_tok_s_per_user": 153.94425844161916,
      "delta_percent": 1.9924450948711758
    },
    {
      "config_id": "eagle4-isolated-frontend-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 8,
      "valid_runs": 1,
      "median_raw_span_tok_s": 726.1461765457339,
      "observed_range_tok_s": [
        726.1461765457339,
        726.1461765457339
      ],
      "median_run_ttft_p95_ms": 710.00470515,
      "median_run_e2e_p95_ms": 11974.5002529,
      "median_run_decode_tok_s_per_user": 97.10438412250501,
      "delta_percent": -27.469646425611703
    },
    {
      "config_id": "eagle4-isolated-frontend-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 16,
      "valid_runs": 1,
      "median_raw_span_tok_s": 733.6978945176152,
      "observed_range_tok_s": [
        733.6978945176152,
        733.6978945176152
      ],
      "median_run_ttft_p95_ms": 12020.099084,
      "median_run_e2e_p95_ms": 23257.751634,
      "median_run_decode_tok_s_per_user": 95.70163598918275,
      "delta_percent": -26.4586992546081
    },
    {
      "config_id": "eagle4-isolated-frontend-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 32,
      "valid_runs": 1,
      "median_raw_span_tok_s": 733.4900239726819,
      "observed_range_tok_s": [
        733.4900239726819,
        733.4900239726819
      ],
      "median_run_ttft_p95_ms": 34536.7252099,
      "median_run_e2e_p95_ms": 45878.59051435,
      "median_run_decode_tok_s_per_user": 95.18412469162458,
      "delta_percent": -26.79104880751295
    },
    {
      "config_id": "maxseq16",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 1,
      "valid_runs": 1,
      "median_raw_span_tok_s": 145.7897295944711,
      "observed_range_tok_s": [
        145.7897295944711,
        145.7897295944711
      ],
      "median_run_ttft_p95_ms": 433.3580050499999,
      "median_run_e2e_p95_ms": 8295.90202385,
      "median_run_decode_tok_s_per_user": 154.2814011036795,
      "delta_percent": 0.7947702759840025
    },
    {
      "config_id": "maxseq16",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 8,
      "valid_runs": 1,
      "median_raw_span_tok_s": 1012.5282175850703,
      "observed_range_tok_s": [
        1012.5282175850703,
        1012.5282175850703
      ],
      "median_run_ttft_p95_ms": 547.6422465499998,
      "median_run_e2e_p95_ms": 9679.15171,
      "median_run_decode_tok_s_per_user": 136.23683422906313,
      "delta_percent": 1.135325086800898
    },
    {
      "config_id": "maxseq16",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 16,
      "valid_runs": 2,
      "median_raw_span_tok_s": 1576.3346982692578,
      "observed_range_tok_s": [
        1543.6626995336858,
        1609.0066970048301
      ],
      "median_run_ttft_p95_ms": 1046.1719742999999,
      "median_run_e2e_p95_ms": 12631.812848449998,
      "median_run_decode_tok_s_per_user": 107.63799246280882,
      "delta_percent": 58.00195828152652
    },
    {
      "config_id": "maxseq16",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 32,
      "valid_runs": 2,
      "median_raw_span_tok_s": 1587.6298302199762,
      "observed_range_tok_s": [
        1562.2251741728444,
        1613.0344862671077
      ],
      "median_run_ttft_p95_ms": 11610.950270649999,
      "median_run_e2e_p95_ms": 22895.09446865,
      "median_run_decode_tok_s_per_user": 106.57691989930734,
      "delta_percent": 58.45984397006543
    },
    {
      "config_id": "sglang-target-api-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 1,
      "valid_runs": 1,
      "median_raw_span_tok_s": 67.10921641236976,
      "observed_range_tok_s": [
        67.10921641236976,
        67.10921641236976
      ],
      "median_run_ttft_p95_ms": 409.62379109999995,
      "median_run_e2e_p95_ms": 15329.9947275,
      "median_run_decode_tok_s_per_user": 68.56549970201472,
      "delta_percent": -53.60264354353647
    },
    {
      "config_id": "sglang-target-api-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 8,
      "valid_runs": 1,
      "median_raw_span_tok_s": 449.5587695426376,
      "observed_range_tok_s": [
        449.5587695426376,
        449.5587695426376
      ],
      "median_run_ttft_p95_ms": 861.4392967499999,
      "median_run_e2e_p95_ms": 18263.76092925,
      "median_run_decode_tok_s_per_user": 58.36076132647327,
      "delta_percent": -55.09629113176109
    },
    {
      "config_id": "sglang-target-api-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 16,
      "valid_runs": 1,
      "median_raw_span_tok_s": 463.3195127778992,
      "observed_range_tok_s": [
        463.3195127778992,
        463.3195127778992
      ],
      "median_run_ttft_p95_ms": 18927.3505341,
      "median_run_e2e_p95_ms": 36348.06638305,
      "median_run_decode_tok_s_per_user": 58.75910070125392,
      "delta_percent": -53.559741843323664
    },
    {
      "config_id": "sglang-target-api-full",
      "input_target": 1024,
      "output_tokens": 1024,
      "concurrency": 32,
      "valid_runs": 1,
      "median_raw_span_tok_s": 457.02471869122076,
      "observed_range_tok_s": [
        457.02471869122076,
        457.02471869122076
      ],
      "median_run_ttft_p95_ms": 54533.537891550004,
      "median_run_e2e_p95_ms": 71959.09729935,
      "median_run_decode_tok_s_per_user": 58.705734999608765,
      "delta_percent": -54.38479157056426
    }
  ],
  "configurations": [
    {
      "config_id": "baseline-mtp3-recheck",
      "config_sha": "cba63bfdce522de45d5fef39151b784613a143106a869918dae8c11699232715",
      "runtime_sha": "7bda4493758cc3bd5d441f866a194751881d7b29dff91467187ea828d909626c",
      "setup_sha": null,
      "serve_sha": "6064a686c080a4e8690be66c6185d191497c807f53ef67e2cae61791b626685a",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "nvcr.io/nvidia/ai-dynamo/vllm-runtime-nightly:20261018-3763cee@sha256:197284c3fd3037d4301ffa8282ed330446a5a371c0223fe1da448529930f064b",
      "framework_source_patch": false,
      "pilot_status": "no full pilot",
      "api_passed": true,
      "valid_capacity_points": 4,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "frontend_args": [
          "--http-host",
          "127.0.0.1",
          "--http-port",
          "8000",
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--dyn-chat-processor",
          "vllm",
          "--reasoning-parser",
          "qwen3",
          "--enable-auto-tool-choice",
          "--tool-call-parser",
          "qwen3_coder"
        ],
        "worker_args": [
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--tensor-parallel-size",
          "1",
          "--max-model-len",
          "131072",
          "--max-num-seqs",
          "8",
          "--gpu-memory-utilization",
          "0.85",
          "--language-model-only",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "fp8",
          "--max-num-batched-tokens",
          "8192",
          "--mamba-cache-mode",
          "align",
          "--speculative-config",
          "{\"method\":\"mtp\",\"num_speculative_tokens\":3}",
          "--reasoning-parser",
          "qwen3"
        ],
        "structured_output_default_thinking": false
      }
    },
    {
      "config_id": "eagle4-depth6-full",
      "config_sha": "34c015b081a4ca2d89435952ae5c8908c05b66d3e0e75b26ec7c9e271403278d",
      "runtime_sha": "905e792039e2b2e3a13796532af7184dc85f14da009e9dc496f4823db563b2af",
      "setup_sha": "4ea300ef9bbc69d13005889a5321d015f070d3662c275e9288c6eee6dafc4e26",
      "serve_sha": "ca8f40eeeb662a46a1a7e7438d5f5a4af278e5babf065e11263a8c9f6174a51e",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "nvcr.io/nvidia/ai-dynamo/vllm-runtime-nightly:20261018-3763cee@sha256:197284c3fd3037d4301ffa8282ed330446a5a371c0223fe1da448529930f064b",
      "framework_source_patch": false,
      "pilot_status": "complete",
      "api_passed": false,
      "valid_capacity_points": 4,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "frontend_args": [
          "--http-host",
          "127.0.0.1",
          "--http-port",
          "8000",
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--dyn-chat-processor",
          "vllm",
          "--reasoning-parser",
          "qwen3",
          "--enable-auto-tool-choice",
          "--tool-call-parser",
          "qwen3_coder"
        ],
        "worker_args": [
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--tensor-parallel-size",
          "1",
          "--max-model-len",
          "131072",
          "--max-num-seqs",
          "8",
          "--gpu-memory-utilization",
          "0.85",
          "--language-model-only",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "auto",
          "--max-num-batched-tokens",
          "8192",
          "--mamba-cache-mode",
          "align",
          "--speculative-config",
          "{\"method\": \"eagle_harness\", \"model\": \"/workspace/eagle-draft\", \"method_config\": {\"mode\": \"static\", \"node_top_k\": 4, \"tree_depth\": 6, \"max_verify_width\": 8, \"depth_stop_gain\": 0.0}}",
          "--attention-backend",
          "CUSTOM",
          "--no-enable-prefix-caching"
        ]
      }
    },
    {
      "config_id": "eagle4-depth6-prefix-full",
      "config_sha": "532f6a5f9463eb78a08c2aca77c564ad5837ba77960bea4333a3e57940aad018",
      "runtime_sha": "905e792039e2b2e3a13796532af7184dc85f14da009e9dc496f4823db563b2af",
      "setup_sha": "4ea300ef9bbc69d13005889a5321d015f070d3662c275e9288c6eee6dafc4e26",
      "serve_sha": "6064a686c080a4e8690be66c6185d191497c807f53ef67e2cae61791b626685a",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "nvcr.io/nvidia/ai-dynamo/vllm-runtime-nightly:20261018-3763cee@sha256:197284c3fd3037d4301ffa8282ed330446a5a371c0223fe1da448529930f064b",
      "framework_source_patch": false,
      "pilot_status": "complete",
      "api_passed": true,
      "valid_capacity_points": 2,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "frontend_args": [
          "--http-host",
          "127.0.0.1",
          "--http-port",
          "8000",
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--dyn-chat-processor",
          "vllm",
          "--reasoning-parser",
          "qwen3",
          "--enable-auto-tool-choice",
          "--tool-call-parser",
          "qwen3_coder"
        ],
        "worker_args": [
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--tensor-parallel-size",
          "1",
          "--max-model-len",
          "131072",
          "--max-num-seqs",
          "8",
          "--gpu-memory-utilization",
          "0.85",
          "--language-model-only",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "auto",
          "--max-num-batched-tokens",
          "8192",
          "--mamba-cache-mode",
          "align",
          "--speculative-config",
          "{\"method\": \"eagle_harness\", \"model\": \"/workspace/eagle-draft\", \"method_config\": {\"mode\": \"static\", \"node_top_k\": 4, \"tree_depth\": 6, \"max_verify_width\": 8, \"depth_stop_gain\": 0.0}}",
          "--attention-backend",
          "CUSTOM",
          "--enable-prefix-caching",
          "--reasoning-parser",
          "qwen3"
        ],
        "structured_output_default_thinking": false
      }
    },
    {
      "config_id": "eagle4-depth6-reasoning-full",
      "config_sha": "53c474de851b9511204e922d27355ba101614ea6192d0362b4793568dbe398f8",
      "runtime_sha": "905e792039e2b2e3a13796532af7184dc85f14da009e9dc496f4823db563b2af",
      "setup_sha": "4ea300ef9bbc69d13005889a5321d015f070d3662c275e9288c6eee6dafc4e26",
      "serve_sha": "ca8f40eeeb662a46a1a7e7438d5f5a4af278e5babf065e11263a8c9f6174a51e",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "nvcr.io/nvidia/ai-dynamo/vllm-runtime-nightly:20261018-3763cee@sha256:197284c3fd3037d4301ffa8282ed330446a5a371c0223fe1da448529930f064b",
      "framework_source_patch": false,
      "pilot_status": "complete",
      "api_passed": false,
      "valid_capacity_points": 1,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "frontend_args": [
          "--http-host",
          "127.0.0.1",
          "--http-port",
          "8000",
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--dyn-chat-processor",
          "vllm",
          "--reasoning-parser",
          "qwen3",
          "--enable-auto-tool-choice",
          "--tool-call-parser",
          "qwen3_coder"
        ],
        "worker_args": [
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--tensor-parallel-size",
          "1",
          "--max-model-len",
          "131072",
          "--max-num-seqs",
          "8",
          "--gpu-memory-utilization",
          "0.85",
          "--language-model-only",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "auto",
          "--max-num-batched-tokens",
          "8192",
          "--mamba-cache-mode",
          "align",
          "--speculative-config",
          "{\"method\": \"eagle_harness\", \"model\": \"/workspace/eagle-draft\", \"method_config\": {\"mode\": \"static\", \"node_top_k\": 4, \"tree_depth\": 6, \"max_verify_width\": 8, \"depth_stop_gain\": 0.0}}",
          "--attention-backend",
          "CUSTOM",
          "--no-enable-prefix-caching",
          "--reasoning-parser",
          "qwen3"
        ]
      }
    },
    {
      "config_id": "eagle4-depth6-structured-full",
      "config_sha": "5ebd8eb1ed701ee9abff71b1e5a5d96a95e54d9d475aa0d3e4f3c4d64a1cc4b3",
      "runtime_sha": "905e792039e2b2e3a13796532af7184dc85f14da009e9dc496f4823db563b2af",
      "setup_sha": "4ea300ef9bbc69d13005889a5321d015f070d3662c275e9288c6eee6dafc4e26",
      "serve_sha": "6064a686c080a4e8690be66c6185d191497c807f53ef67e2cae61791b626685a",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "nvcr.io/nvidia/ai-dynamo/vllm-runtime-nightly:20261018-3763cee@sha256:197284c3fd3037d4301ffa8282ed330446a5a371c0223fe1da448529930f064b",
      "framework_source_patch": false,
      "pilot_status": "complete",
      "api_passed": true,
      "valid_capacity_points": 1,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "frontend_args": [
          "--http-host",
          "127.0.0.1",
          "--http-port",
          "8000",
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--dyn-chat-processor",
          "vllm",
          "--reasoning-parser",
          "qwen3",
          "--enable-auto-tool-choice",
          "--tool-call-parser",
          "qwen3_coder"
        ],
        "worker_args": [
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--tensor-parallel-size",
          "1",
          "--max-model-len",
          "131072",
          "--max-num-seqs",
          "8",
          "--gpu-memory-utilization",
          "0.85",
          "--language-model-only",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "auto",
          "--max-num-batched-tokens",
          "8192",
          "--mamba-cache-mode",
          "align",
          "--speculative-config",
          "{\"method\": \"eagle_harness\", \"model\": \"/workspace/eagle-draft\", \"method_config\": {\"mode\": \"static\", \"node_top_k\": 4, \"tree_depth\": 6, \"max_verify_width\": 8, \"depth_stop_gain\": 0.0}}",
          "--attention-backend",
          "CUSTOM",
          "--no-enable-prefix-caching",
          "--reasoning-parser",
          "qwen3"
        ],
        "structured_output_default_thinking": false
      }
    },
    {
      "config_id": "eagle4-integration-4k",
      "config_sha": "e2446877a797235004889adbc4e61a207eb3c847c42cf2eb87286b33b62b1c42",
      "runtime_sha": "905e792039e2b2e3a13796532af7184dc85f14da009e9dc496f4823db563b2af",
      "setup_sha": "e17de8c931c26effd4ac8f3d737e6b7d92c483a7f5d04d2e6aed3d273af0d5ef",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "nvcr.io/nvidia/ai-dynamo/vllm-runtime-nightly:20261018-3763cee@sha256:197284c3fd3037d4301ffa8282ed330446a5a371c0223fe1da448529930f064b",
      "framework_source_patch": false,
      "pilot_status": "startup_failed",
      "api_passed": null,
      "valid_capacity_points": 0,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "frontend_args": [
          "--http-host",
          "127.0.0.1",
          "--http-port",
          "8000",
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--dyn-chat-processor",
          "vllm",
          "--reasoning-parser",
          "qwen3",
          "--enable-auto-tool-choice",
          "--tool-call-parser",
          "qwen3_coder"
        ],
        "worker_args": [
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--tensor-parallel-size",
          "1",
          "--max-model-len",
          "4096",
          "--max-num-seqs",
          "1",
          "--gpu-memory-utilization",
          "0.85",
          "--language-model-only",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "auto",
          "--max-num-batched-tokens",
          "8192",
          "--mamba-cache-mode",
          "align",
          "--speculative-config",
          "{\"method\": \"eagle_harness\", \"model\": \"/workspace/eagle-draft\", \"method_config\": {\"mode\": \"static\", \"node_top_k\": 4, \"tree_depth\": 2, \"max_verify_width\": 8, \"depth_stop_gain\": 0.0}}",
          "--attention-backend",
          "CUSTOM",
          "--no-enable-prefix-caching"
        ]
      }
    },
    {
      "config_id": "eagle4-isolated-frontend-4k",
      "config_sha": "e2446877a797235004889adbc4e61a207eb3c847c42cf2eb87286b33b62b1c42",
      "runtime_sha": "905e792039e2b2e3a13796532af7184dc85f14da009e9dc496f4823db563b2af",
      "setup_sha": "e17de8c931c26effd4ac8f3d737e6b7d92c483a7f5d04d2e6aed3d273af0d5ef",
      "serve_sha": "ca8f40eeeb662a46a1a7e7438d5f5a4af278e5babf065e11263a8c9f6174a51e",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "nvcr.io/nvidia/ai-dynamo/vllm-runtime-nightly:20261018-3763cee@sha256:197284c3fd3037d4301ffa8282ed330446a5a371c0223fe1da448529930f064b",
      "framework_source_patch": false,
      "pilot_status": "no full pilot",
      "api_passed": null,
      "valid_capacity_points": 0,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "frontend_args": [
          "--http-host",
          "127.0.0.1",
          "--http-port",
          "8000",
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--dyn-chat-processor",
          "vllm",
          "--reasoning-parser",
          "qwen3",
          "--enable-auto-tool-choice",
          "--tool-call-parser",
          "qwen3_coder"
        ],
        "worker_args": [
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--tensor-parallel-size",
          "1",
          "--max-model-len",
          "4096",
          "--max-num-seqs",
          "1",
          "--gpu-memory-utilization",
          "0.85",
          "--language-model-only",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "auto",
          "--max-num-batched-tokens",
          "8192",
          "--mamba-cache-mode",
          "align",
          "--speculative-config",
          "{\"method\": \"eagle_harness\", \"model\": \"/workspace/eagle-draft\", \"method_config\": {\"mode\": \"static\", \"node_top_k\": 4, \"tree_depth\": 2, \"max_verify_width\": 8, \"depth_stop_gain\": 0.0}}",
          "--attention-backend",
          "CUSTOM",
          "--no-enable-prefix-caching"
        ]
      }
    },
    {
      "config_id": "eagle4-isolated-frontend-full",
      "config_sha": "0c1d02d1d167852046a5a1e0ca7c9ba38b829438b2378b1854252f85639a4004",
      "runtime_sha": "905e792039e2b2e3a13796532af7184dc85f14da009e9dc496f4823db563b2af",
      "setup_sha": "e17de8c931c26effd4ac8f3d737e6b7d92c483a7f5d04d2e6aed3d273af0d5ef",
      "serve_sha": "ca8f40eeeb662a46a1a7e7438d5f5a4af278e5babf065e11263a8c9f6174a51e",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "nvcr.io/nvidia/ai-dynamo/vllm-runtime-nightly:20261018-3763cee@sha256:197284c3fd3037d4301ffa8282ed330446a5a371c0223fe1da448529930f064b",
      "framework_source_patch": false,
      "pilot_status": "complete",
      "api_passed": null,
      "valid_capacity_points": 4,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "frontend_args": [
          "--http-host",
          "127.0.0.1",
          "--http-port",
          "8000",
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--dyn-chat-processor",
          "vllm",
          "--reasoning-parser",
          "qwen3",
          "--enable-auto-tool-choice",
          "--tool-call-parser",
          "qwen3_coder"
        ],
        "worker_args": [
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--tensor-parallel-size",
          "1",
          "--max-model-len",
          "131072",
          "--max-num-seqs",
          "8",
          "--gpu-memory-utilization",
          "0.85",
          "--language-model-only",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "auto",
          "--max-num-batched-tokens",
          "8192",
          "--mamba-cache-mode",
          "align",
          "--speculative-config",
          "{\"method\": \"eagle_harness\", \"model\": \"/workspace/eagle-draft\", \"method_config\": {\"mode\": \"static\", \"node_top_k\": 4, \"tree_depth\": 2, \"max_verify_width\": 8, \"depth_stop_gain\": 0.0}}",
          "--attention-backend",
          "CUSTOM",
          "--no-enable-prefix-caching"
        ]
      }
    },
    {
      "config_id": "maxseq16",
      "config_sha": "1aceda12bac158c5c475416bbd0acd08b458cc0e25ac230acb60b4c04224c9ca",
      "runtime_sha": "7bda4493758cc3bd5d441f866a194751881d7b29dff91467187ea828d909626c",
      "setup_sha": null,
      "serve_sha": "6064a686c080a4e8690be66c6185d191497c807f53ef67e2cae61791b626685a",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "nvcr.io/nvidia/ai-dynamo/vllm-runtime-nightly:20261018-3763cee@sha256:197284c3fd3037d4301ffa8282ed330446a5a371c0223fe1da448529930f064b",
      "framework_source_patch": false,
      "pilot_status": "complete",
      "api_passed": true,
      "valid_capacity_points": 6,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "frontend_args": [
          "--http-host",
          "127.0.0.1",
          "--http-port",
          "8000",
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--dyn-chat-processor",
          "vllm",
          "--reasoning-parser",
          "qwen3",
          "--enable-auto-tool-choice",
          "--tool-call-parser",
          "qwen3_coder"
        ],
        "worker_args": [
          "--discovery-backend",
          "file",
          "--request-plane",
          "tcp",
          "--event-plane",
          "zmq",
          "--tensor-parallel-size",
          "1",
          "--max-model-len",
          "131072",
          "--max-num-seqs",
          "16",
          "--gpu-memory-utilization",
          "0.85",
          "--language-model-only",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "fp8",
          "--max-num-batched-tokens",
          "8192",
          "--mamba-cache-mode",
          "align",
          "--speculative-config",
          "{\"method\":\"mtp\",\"num_speculative_tokens\":3}",
          "--reasoning-parser",
          "qwen3"
        ],
        "structured_output_default_thinking": false
      }
    },
    {
      "config_id": "openjet-unsloth-udq4xl-mtp4",
      "config_sha": "85a075eef6c81e884ce296c01cf9ca7c08e4907b3ce632e243e9d6884501b766",
      "runtime_sha": "50e9759a278e135352111d7ed56f36378a3d91a415a77581b45bce84f8252a25",
      "setup_sha": "29d338fc16042f27d8b44dacbe73c186211821b56fd5a1a8ede6d105f2780d5c",
      "serve_sha": "9dab416b44291bde2b3e0f00a5a3d696b13a643dccf95e88fdda793d107f8ba2",
      "model": "unsloth/Qwen3.8-27B-GGUF",
      "model_revision": "4ca720788d1e01f1bff70c033e0d0028fd02e502",
      "image": "lmsysorg/sglang@sha256:5fc09edb6a0bb1d5b6992b543723c66d08f86d7a507e951e87e0d05db4ac766e",
      "framework_source_patch": false,
      "pilot_status": "no full pilot",
      "api_passed": null,
      "valid_capacity_points": 0,
      "serving": {
        "model": "unsloth/Qwen3.8-27B-GGUF",
        "revision": "4ca720788d1e01f1bff70c033e0d0028fd02e502",
        "served_model": "Qwen3.8-27B-UD-Q4_K_XL.gguf",
        "framework": "openjet",
        "frontend_args": [],
        "worker_args": [
          "--host",
          "127.0.0.1",
          "--port",
          "8000",
          "--ctx",
          "131072",
          "--draft",
          "4",
          "--kv",
          "f16"
        ],
        "model_file": "Qwen3.8-27B-UD-Q4_K_XL.gguf",
        "model_sha256": "3f227079003add2511437e5b1e94812e363385225bf6a9b47b0054a72bc8b01e"
      }
    },
    {
      "config_id": "openjet-unsloth-udq4xl-mtp4-retry-20261022",
      "config_sha": "85a075eef6c81e884ce296c01cf9ca7c08e4907b3ce632e243e9d6884501b766",
      "runtime_sha": "03dddcb90bf56df929758d34116b127141caf193afe83ef001642ba35c442fca",
      "setup_sha": "29d338fc16042f27d8b44dacbe73c186211821b56fd5a1a8ede6d105f2780d5c",
      "serve_sha": "9dab416b44291bde2b3e0f00a5a3d696b13a643dccf95e88fdda793d107f8ba2",
      "model": "unsloth/Qwen3.8-27B-GGUF",
      "model_revision": "4ca720788d1e01f1bff70c033e0d0028fd02e502",
      "image": "lmsysorg/sglang@sha256:5fc09edb6a0bb1d5b6992b543723c66d08f86d7a507e951e87e0d05db4ac766e",
      "framework_source_patch": false,
      "pilot_status": "standalone-trial-complete",
      "api_passed": null,
      "valid_capacity_points": 0,
      "serving": {
        "model": "unsloth/Qwen3.8-27B-GGUF",
        "revision": "4ca720788d1e01f1bff70c033e0d0028fd02e502",
        "served_model": "Qwen3.8-27B-UD-Q4_K_XL.gguf",
        "framework": "openjet",
        "frontend_args": [],
        "worker_args": [
          "--host",
          "127.0.0.1",
          "--port",
          "8000",
          "--ctx",
          "131072",
          "--draft",
          "4",
          "--kv",
          "f16"
        ],
        "model_file": "Qwen3.8-27B-UD-Q4_K_XL.gguf",
        "model_sha256": "3f227079003add2511437e5b1e94812e363385225bf6a9b47b0054a72bc8b01e"
      }
    },
    {
      "config_id": "sglang-dflash2-api-4k",
      "config_sha": "9bb068f917fbd396b9d0b741bf9f8f59ecb093ad04ace2ec6f602b4aa8a3c968",
      "runtime_sha": "59289d2af0278031437dec2713719f2062476445b8ceb8e16af75fbd5ec6b905",
      "setup_sha": "32405077d9194e34795197a2ade3d12b64133e5c0fabe780a5f31ff7b610203a",
      "serve_sha": "75f9dad11c535bed7889e57f098a99589616655ed274f2932310b5e77db000da",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "lmsysorg/sglang@sha256:5fc09edb6a0bb1d5b6992b543723c66d08f86d7a507e951e87e0d05db4ac766e",
      "framework_source_patch": true,
      "pilot_status": "complete",
      "api_passed": null,
      "valid_capacity_points": 0,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "framework": "sglang",
        "frontend_args": [],
        "worker_args": [
          "--host",
          "127.0.0.1",
          "--port",
          "8000",
          "--tp-size",
          "1",
          "--context-length",
          "4096",
          "--max-running-requests",
          "1",
          "--mem-fraction-static",
          "0.75",
          "--chunked-prefill-size",
          "8192",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "bf16",
          "--speculative-algorithm",
          "DFLASH",
          "--speculative-draft-model-path",
          "/workspace/dflash2-draft",
          "--speculative-draft-model-revision",
          "50307d4c4cde6860d4eee73e2547cd786fe8e8a4",
          "--speculative-num-draft-tokens",
          "8",
          "--speculative-draft-kv-cache-dtype",
          "bf16",
          "--enable-torch-compile",
          "--reasoning-parser",
          "qwen3",
          "--tool-call-parser",
          "qwen3_coder",
          "--enable-metrics",
          "--speculative-draft-model-quantization",
          "unquant"
        ],
        "structured_output_default_thinking": false
      }
    },
    {
      "config_id": "sglang-dflash2-api-full",
      "config_sha": "7de6eb9823288f39fc48a486596ba6c32f5c9d61273146072575736e9f9221e0",
      "runtime_sha": "59289d2af0278031437dec2713719f2062476445b8ceb8e16af75fbd5ec6b905",
      "setup_sha": "32405077d9194e34795197a2ade3d12b64133e5c0fabe780a5f31ff7b610203a",
      "serve_sha": "75f9dad11c535bed7889e57f098a99589616655ed274f2932310b5e77db000da",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "lmsysorg/sglang@sha256:5fc09edb6a0bb1d5b6992b543723c66d08f86d7a507e951e87e0d05db4ac766e",
      "framework_source_patch": true,
      "pilot_status": "failed",
      "api_passed": true,
      "valid_capacity_points": 0,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "framework": "sglang",
        "frontend_args": [],
        "worker_args": [
          "--host",
          "127.0.0.1",
          "--port",
          "8000",
          "--tp-size",
          "1",
          "--context-length",
          "131072",
          "--max-running-requests",
          "8",
          "--mem-fraction-static",
          "0.75",
          "--chunked-prefill-size",
          "8192",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "bf16",
          "--speculative-algorithm",
          "DFLASH",
          "--speculative-draft-model-path",
          "/workspace/dflash2-draft",
          "--speculative-draft-model-revision",
          "50307d4c4cde6860d4eee73e2547cd786fe8e8a4",
          "--speculative-num-draft-tokens",
          "8",
          "--speculative-draft-kv-cache-dtype",
          "bf16",
          "--enable-torch-compile",
          "--reasoning-parser",
          "qwen3",
          "--tool-call-parser",
          "qwen3_coder",
          "--enable-metrics",
          "--speculative-draft-model-quantization",
          "unquant"
        ],
        "structured_output_default_thinking": false
      }
    },
    {
      "config_id": "sglang-target-api-full",
      "config_sha": "5948fc9fce19cfdfeeac37a7bf7b8a984962e66b2dd94c0c49ddffb19077370e",
      "runtime_sha": "59289d2af0278031437dec2713719f2062476445b8ceb8e16af75fbd5ec6b905",
      "setup_sha": "af13f7fe357e9c4384721375fba9b747e7d16c3941259b4a40ea2a8c419fe89f",
      "serve_sha": "75f9dad11c535bed7889e57f098a99589616655ed274f2932310b5e77db000da",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "lmsysorg/sglang@sha256:5fc09edb6a0bb1d5b6992b543723c66d08f86d7a507e951e87e0d05db4ac766e",
      "framework_source_patch": true,
      "pilot_status": "complete",
      "api_passed": true,
      "valid_capacity_points": 4,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "framework": "sglang",
        "frontend_args": [],
        "worker_args": [
          "--host",
          "127.0.0.1",
          "--port",
          "8000",
          "--tp-size",
          "1",
          "--context-length",
          "131072",
          "--max-running-requests",
          "8",
          "--mem-fraction-static",
          "0.75",
          "--chunked-prefill-size",
          "8192",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "bf16",
          "--enable-torch-compile",
          "--reasoning-parser",
          "qwen3",
          "--tool-call-parser",
          "qwen3_coder",
          "--enable-metrics"
        ],
        "structured_output_default_thinking": false
      }
    },
    {
      "config_id": "sglang-target-linear-eager-full",
      "config_sha": "5948fc9fce19cfdfeeac37a7bf7b8a984962e66b2dd94c0c49ddffb19077370e",
      "runtime_sha": "59289d2af0278031437dec2713719f2062476445b8ceb8e16af75fbd5ec6b905",
      "setup_sha": "af13f7fe357e9c4384721375fba9b747e7d16c3941259b4a40ea2a8c419fe89f",
      "serve_sha": "09f4449cb314b339fb7762a02c534a14190e3bc2e69011cc8a6a8f85ab2fe983",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "lmsysorg/sglang@sha256:5fc09edb6a0bb1d5b6992b543723c66d08f86d7a507e951e87e0d05db4ac766e",
      "framework_source_patch": true,
      "pilot_status": "failed",
      "api_passed": false,
      "valid_capacity_points": 0,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "framework": "sglang",
        "frontend_args": [],
        "worker_args": [
          "--host",
          "127.0.0.1",
          "--port",
          "8000",
          "--tp-size",
          "1",
          "--context-length",
          "131072",
          "--max-running-requests",
          "8",
          "--mem-fraction-static",
          "0.75",
          "--chunked-prefill-size",
          "8192",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "bf16",
          "--enable-torch-compile",
          "--reasoning-parser",
          "qwen3",
          "--tool-call-parser",
          "qwen3_coder",
          "--enable-metrics"
        ],
        "structured_output_default_thinking": false
      }
    },
    {
      "config_id": "sglang-target-native-full",
      "config_sha": "37f2327d97c7952c5bec5e64fb7235d7ab100e1097081ba4881182866d0e239b",
      "runtime_sha": "59289d2af0278031437dec2713719f2062476445b8ceb8e16af75fbd5ec6b905",
      "setup_sha": "50ef8f6e745e13a27c04bd0fa947f7245ba5dde8566a8b294738c89cba772517",
      "serve_sha": "09f4449cb314b339fb7762a02c534a14190e3bc2e69011cc8a6a8f85ab2fe983",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "lmsysorg/sglang@sha256:5fc09edb6a0bb1d5b6992b543723c66d08f86d7a507e951e87e0d05db4ac766e",
      "framework_source_patch": false,
      "pilot_status": "failed",
      "api_passed": null,
      "valid_capacity_points": 0,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "framework": "sglang",
        "frontend_args": [],
        "worker_args": [
          "--host",
          "127.0.0.1",
          "--port",
          "8000",
          "--tp-size",
          "1",
          "--context-length",
          "131072",
          "--max-running-requests",
          "8",
          "--mem-fraction-static",
          "0.85",
          "--chunked-prefill-size",
          "8192",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "bf16",
          "--enable-torch-compile",
          "--reasoning-parser",
          "qwen3",
          "--tool-call-parser",
          "qwen3_coder",
          "--enable-metrics"
        ],
        "structured_output_default_thinking": false
      }
    },
    {
      "config_id": "sglang-target-only-full",
      "config_sha": "3e9881c2456f03acd75d249e6002144198d13fcbd797ed702a6924dfef1ad1db",
      "runtime_sha": "59289d2af0278031437dec2713719f2062476445b8ceb8e16af75fbd5ec6b905",
      "setup_sha": "50ef8f6e745e13a27c04bd0fa947f7245ba5dde8566a8b294738c89cba772517",
      "serve_sha": "09f4449cb314b339fb7762a02c534a14190e3bc2e69011cc8a6a8f85ab2fe983",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "lmsysorg/sglang@sha256:5fc09edb6a0bb1d5b6992b543723c66d08f86d7a507e951e87e0d05db4ac766e",
      "framework_source_patch": false,
      "pilot_status": "failed",
      "api_passed": null,
      "valid_capacity_points": 0,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "framework": "sglang",
        "frontend_args": [],
        "worker_args": [
          "--host",
          "127.0.0.1",
          "--port",
          "8000",
          "--tp-size",
          "1",
          "--context-length",
          "131072",
          "--max-running-requests",
          "8",
          "--mem-fraction-static",
          "0.85",
          "--chunked-prefill-size",
          "8192",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "bf16",
          "--enable-torch-compile",
          "--language-model-only",
          "--reasoning-parser",
          "qwen3",
          "--tool-call-parser",
          "qwen3_coder",
          "--enable-metrics"
        ],
        "structured_output_default_thinking": false
      }
    },
    {
      "config_id": "sglang-target-room75-full",
      "config_sha": "5948fc9fce19cfdfeeac37a7bf7b8a984962e66b2dd94c0c49ddffb19077370e",
      "runtime_sha": "59289d2af0278031437dec2713719f2062476445b8ceb8e16af75fbd5ec6b905",
      "setup_sha": "50ef8f6e745e13a27c04bd0fa947f7245ba5dde8566a8b294738c89cba772517",
      "serve_sha": "09f4449cb314b339fb7762a02c534a14190e3bc2e69011cc8a6a8f85ab2fe983",
      "model": "nvidia/Qwen3.8-27B-NVFP4",
      "model_revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
      "image": "lmsysorg/sglang@sha256:5fc09edb6a0bb1d5b6992b543723c66d08f86d7a507e951e87e0d05db4ac766e",
      "framework_source_patch": false,
      "pilot_status": "operator-aborted-memory-retries",
      "api_passed": null,
      "valid_capacity_points": 0,
      "serving": {
        "model": "nvidia/Qwen3.8-27B-NVFP4",
        "revision": "482ca0f3832238542f8f5295dde86b5f22711d80",
        "served_model": "Qwen/Qwen3.8-27B-NVFP4",
        "framework": "sglang",
        "frontend_args": [],
        "worker_args": [
          "--host",
          "127.0.0.1",
          "--port",
          "8000",
          "--tp-size",
          "1",
          "--context-length",
          "131072",
          "--max-running-requests",
          "8",
          "--mem-fraction-static",
          "0.75",
          "--chunked-prefill-size",
          "8192",
          "--dtype",
          "bfloat16",
          "--kv-cache-dtype",
          "bf16",
          "--enable-torch-compile",
          "--reasoning-parser",
          "qwen3",
          "--tool-call-parser",
          "qwen3_coder",
          "--enable-metrics"
        ],
        "structured_output_default_thinking": false
      }
    }
  ]
}
