{ "profile": "l4", "slo": { "ttft_ms": 1000.0, "tpot_ms": 100.0 }, "environment": { "sglang": "0.5.22", "sglang_kernel": "0.4.9", "flashinfer": "0.7.1", "torch": "2.14.1+cu130", "cuda": "13.0", "driver": "615.71.09\n615.71.09", "gpu": "NVIDIA L4", "compute_capability": "8.9", "gpu_memory_total_gib": 22.1, "gpu_memory_free_gib_at_start": 21.9, "python": "3.12.15", "machine": "x86_64", "profile": "l4", "models": [ "Qwen/Qwen3-8B", "Qwen/Qwen3-8B-FP8", "Tengyunw/qwen3_8b_eagle3" ] }, "load": { "default": { "kv_pool_tokens": 24094, "points": [ { "x": 1.0, "output_throughput": { "median": 10.11, "min": 10.02, "max": 14.9, "n": 3 }, "median_ttft_ms": { "median": 679.37, "min": 183.1, "max": 694.21, "n": 3 }, "p99_ttft_ms": { "median": 701.93, "min": 498.36, "max": 704.52, "n": 3 }, "median_tpot_ms": { "median": 103.38, "min": 60.13, "max": 103.39, "n": 3 }, "p99_tpot_ms": { "median": 103.39, "min": 88.74, "max": 103.4, "n": 3 }, "goodput_rps": { "median": 0.02, "min": 0.02, "max": 0.12, "n": 3 }, "request_throughput": { "median": 0.08, "min": 0.08, "max": 0.12, "n": 3 }, "completed": [ 16, 16, 16 ], "ttft_quantiles_ms": [ 71.0, 71.0, 71.9, 72.2, 169.2, 172.3, 174.9, 176.3, 178.8, 179.1, 182.7, 183.5, 183.7, 185.9, 201.2, 287.8, 298.7, 298.7, 299.7, 456.3, 461.0, 474.1, 480.4, 480.4, 487.6, 494.9, 499.0, 620.0, 642.7, 657.8, 663.3, 691.2, 691.6, 694.1, 694.1, 694.4, 694.5, 694.7, 695.4, 696.2, 696.3, 696.7, 696.9, 697.3, 699.8, 700.6, 701.7, 702.0, 702.4, 703.2, 704.8 ], "tpot_quantiles_ms": [ 60.0, 60.0, 60.0, 60.0, 60.1, 60.1, 60.1, 60.1, 60.1, 60.1, 60.1, 60.2, 60.2, 60.2, 60.3, 65.5, 67.4, 67.4, 69.9, 81.4, 81.7, 85.5, 85.6, 85.7, 88.0, 88.2, 88.8, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4 ] }, { "x": 2.0, "output_throughput": { "median": 25.12, "min": 24.61, "max": 26.76, "n": 3 }, "median_ttft_ms": { "median": 130.15, "min": 124.12, "max": 134.35, "n": 3 }, "p99_ttft_ms": { "median": 204.88, "min": 194.87, "max": 215.77, "n": 3 }, "median_tpot_ms": { "median": 81.17, "min": 72.21, "max": 81.4, "n": 3 }, "p99_tpot_ms": { "median": 87.56, "min": 86.49, "max": 90.78, "n": 3 }, "goodput_rps": { "median": 0.2, "min": 0.19, "max": 0.21, "n": 3 }, "request_throughput": { "median": 0.2, "min": 0.19, "max": 0.21, "n": 3 }, "completed": [ 16, 16, 16 ], "ttft_quantiles_ms": [ 97.5, 97.5, 97.6, 98.0, 98.5, 98.5, 99.1, 99.1, 100.1, 100.1, 101.5, 101.6, 102.6, 103.0, 107.6, 110.5, 111.7, 111.7, 115.1, 115.5, 117.3, 121.1, 122.9, 125.5, 132.8, 133.0, 133.1, 134.8, 135.6, 135.6, 136.0, 136.4, 136.8, 137.7, 137.7, 137.7, 138.2, 139.6, 140.0, 163.8, 182.0, 189.6, 190.4, 193.3, 195.1, 196.1, 201.0, 202.0, 205.4, 211.4, 216.5 ], "tpot_quantiles_ms": [ 63.7, 63.7, 64.0, 64.3, 64.6, 64.9, 65.2, 65.4, 65.7, 67.5, 67.8, 70.6, 71.2, 72.1, 72.4, 73.2, 73.8, 73.8, 76.6, 77.2, 79.4, 79.6, 79.7, 79.9, 80.3, 80.4, 80.8, 81.0, 81.1, 81.3, 81.4, 81.4, 81.6, 81.9, 81.9, 82.0, 82.0, 82.1, 82.7, 84.8, 85.0, 85.9, 86.6, 86.8, 87.0, 87.1, 87.1, 87.4, 87.6, 90.5, 90.8 ] }, { "x": 4.0, "output_throughput": { "median": 57.43, "min": 55.77, "max": 59.31, "n": 3 }, "median_ttft_ms": { "median": 140.68, "min": 139.74, "max": 147.1, "n": 3 }, "p99_ttft_ms": { "median": 150.88, "min": 146.06, "max": 319.92, "n": 3 }, "median_tpot_ms": { "median": 67.2, "min": 66.15, "max": 70.25, "n": 3 }, "p99_tpot_ms": { "median": 71.58, "min": 68.9, "max": 87.86, "n": 3 }, "goodput_rps": { "median": 0.45, "min": 0.44, "max": 0.46, "n": 3 }, "request_throughput": { "median": 0.45, "min": 0.44, "max": 0.46, "n": 3 }, "completed": [ 16, 16, 16 ], "ttft_quantiles_ms": [ 86.5, 86.5, 87.4, 87.9, 89.1, 89.7, 89.8, 91.0, 91.5, 91.9, 92.5, 92.7, 130.3, 137.9, 138.0, 138.4, 138.6, 138.6, 138.7, 139.0, 139.5, 139.9, 140.3, 140.3, 141.1, 141.6, 141.6, 142.1, 142.7, 144.2, 144.4, 144.7, 144.9, 145.0, 145.0, 145.4, 145.5, 146.0, 146.1, 146.9, 147.3, 147.9, 148.6, 149.2, 149.8, 150.3, 150.7, 150.9, 221.7, 319.4, 320.0 ], "tpot_quantiles_ms": [ 64.5, 64.5, 64.5, 64.5, 64.6, 64.6, 64.6, 64.6, 64.6, 64.6, 65.0, 65.0, 65.0, 65.7, 65.7, 65.7, 65.7, 65.7, 65.8, 65.8, 66.1, 66.2, 66.2, 66.2, 66.2, 66.6, 68.2, 68.2, 68.3, 68.5, 68.5, 68.5, 68.7, 68.9, 68.9, 69.0, 69.0, 69.0, 69.4, 71.1, 71.1, 71.1, 71.1, 71.1, 71.1, 71.5, 71.6, 86.5, 86.5, 87.2, 88.0 ] }, { "x": 8.0, "output_throughput": { "median": 93.57, "min": 92.38, "max": 94.45, "n": 3 }, "median_ttft_ms": { "median": 313.36, "min": 283.29, "max": 329.46, "n": 3 }, "p99_ttft_ms": { "median": 3876.19, "min": 3529.34, "max": 3919.08, "n": 3 }, "median_tpot_ms": { "median": 72.09, "min": 71.16, "max": 73.67, "n": 3 }, "p99_tpot_ms": { "median": 98.26, "min": 96.84, "max": 100.17, "n": 3 }, "goodput_rps": { "median": 0.42, "min": 0.38, "max": 0.43, "n": 3 }, "request_throughput": { "median": 0.73, "min": 0.72, "max": 0.74, "n": 3 }, "completed": [ 32, 32, 32 ], "ttft_quantiles_ms": [ 88.3, 88.3, 89.9, 108.1, 140.1, 141.5, 141.8, 142.1, 142.5, 143.0, 143.1, 143.4, 144.0, 145.1, 146.4, 154.4, 155.1, 160.3, 160.9, 161.6, 162.3, 166.9, 167.5, 168.2, 201.1, 204.2, 457.9, 466.2, 647.2, 1227.5, 1370.0, 1371.3, 1375.2, 1873.9, 2031.4, 3526.0, 3526.3, 3526.9, 3527.4, 3528.9, 3605.6, 3605.9, 3606.9, 3789.8, 3790.5, 3791.5, 3872.9, 3874.6, 3876.5, 3917.7, 3919.3 ], "tpot_quantiles_ms": [ 66.2, 66.3, 66.3, 66.3, 66.3, 66.4, 66.6, 66.6, 66.6, 66.6, 66.7, 66.7, 67.2, 70.9, 70.9, 70.9, 70.9, 71.2, 71.2, 71.2, 71.3, 71.3, 71.3, 71.3, 71.7, 72.2, 72.2, 72.5, 72.5, 72.5, 72.8, 72.8, 72.8, 73.2, 73.2, 73.7, 73.7, 73.7, 73.7, 74.7, 74.7, 74.7, 89.0, 89.5, 89.5, 89.7, 90.9, 93.8, 96.9, 97.7, 101.3 ] }, { "x": 16.0, "output_throughput": { "median": 159.13, "min": 157.39, "max": 162.85, "n": 3 }, "median_ttft_ms": { "median": 332.12, "min": 294.81, "max": 340.87, "n": 3 }, "p99_ttft_ms": { "median": 6492.19, "min": 6430.44, "max": 6618.7, "n": 3 }, "median_tpot_ms": { "median": 79.01, "min": 75.41, "max": 80.84, "n": 3 }, "p99_tpot_ms": { "median": 123.29, "min": 117.04, "max": 124.42, "n": 3 }, "goodput_rps": { "median": 0.62, "min": 0.61, "max": 0.64, "n": 3 }, "request_throughput": { "median": 1.24, "min": 1.23, "max": 1.27, "n": 3 }, "completed": [ 64, 64, 64 ], "ttft_quantiles_ms": [ 90.9, 93.8, 152.5, 153.8, 155.7, 158.4, 159.0, 159.3, 159.7, 160.1, 160.5, 161.4, 162.5, 204.6, 206.4, 208.4, 209.1, 209.9, 210.4, 211.1, 211.7, 212.0, 212.3, 212.6, 213.0, 213.9, 471.9, 948.8, 1745.1, 1801.9, 1894.3, 3428.5, 3431.6, 4469.0, 4470.6, 4827.1, 4829.3, 5113.8, 5115.5, 5188.2, 5189.8, 5269.4, 5271.8, 5616.4, 6373.3, 6374.2, 6423.6, 6428.9, 6430.4, 6492.4, 6619.3 ], "tpot_quantiles_ms": [ 70.0, 70.0, 70.0, 70.0, 70.0, 70.1, 70.1, 70.4, 70.4, 70.4, 70.5, 70.5, 71.0, 72.1, 72.1, 75.0, 75.0, 75.0, 75.1, 75.4, 75.4, 75.8, 78.7, 78.7, 79.0, 79.0, 79.3, 79.4, 79.4, 80.8, 80.8, 80.8, 81.2, 84.8, 84.9, 84.9, 85.5, 85.5, 86.9, 86.9, 89.3, 89.3, 91.4, 91.5, 94.2, 94.2, 110.8, 112.8, 113.4, 121.9, 125.8 ] }, { "x": 32.0, "output_throughput": { "median": 194.16, "min": 193.3, "max": 195.69, "n": 3 }, "median_ttft_ms": { "median": 6474.17, "min": 6199.97, "max": 6681.5, "n": 3 }, "p99_ttft_ms": { "median": 11379.32, "min": 11368.48, "max": 11920.12, "n": 3 }, "median_tpot_ms": { "median": 113.18, "min": 109.58, "max": 116.01, "n": 3 }, "p99_tpot_ms": { "median": 173.11, "min": 171.94, "max": 175.44, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.52, "min": 1.51, "max": 1.53, "n": 3 }, "completed": [ 128, 128, 128 ], "ttft_quantiles_ms": [ 110.2, 381.1, 1038.6, 1071.9, 1088.3, 1368.5, 1906.3, 2724.6, 2753.5, 2852.2, 3794.0, 3919.2, 4004.5, 4059.0, 4384.7, 4471.5, 4474.7, 4519.4, 4579.3, 4605.3, 4667.7, 4898.4, 6198.8, 6399.2, 6477.5, 6484.9, 6598.8, 6680.0, 6737.7, 6771.0, 6929.1, 7018.6, 7264.4, 7315.7, 7485.7, 9175.7, 9287.7, 9355.8, 9367.2, 9370.9, 9428.5, 9586.1, 9715.8, 9739.6, 9928.9, 10786.0, 10893.1, 11085.7, 11367.0, 11379.1, 11920.5 ], "tpot_quantiles_ms": [ 82.3, 82.3, 83.8, 84.6, 84.9, 87.3, 88.6, 89.2, 89.6, 91.4, 95.0, 96.5, 96.5, 98.3, 98.5, 98.9, 99.4, 99.9, 101.7, 102.8, 104.3, 105.3, 107.4, 109.6, 113.2, 115.2, 116.0, 118.3, 119.4, 119.5, 121.0, 122.0, 122.6, 124.1, 126.2, 126.6, 127.1, 130.1, 135.4, 137.4, 140.5, 140.7, 142.2, 145.3, 145.4, 146.0, 147.2, 161.7, 164.9, 170.9, 180.5 ] }, { "x": 64.0, "output_throughput": { "median": 199.52, "min": 198.92, "max": 200.39, "n": 3 }, "median_ttft_ms": { "median": 27303.19, "min": 27049.37, "max": 27426.98, "n": 3 }, "p99_ttft_ms": { "median": 36022.18, "min": 35825.14, "max": 36040.4, "n": 3 }, "median_tpot_ms": { "median": 123.91, "min": 123.64, "max": 126.23, "n": 3 }, "p99_tpot_ms": { "median": 172.56, "min": 171.42, "max": 173.32, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.56, "min": 1.55, "max": 1.57, "n": 3 }, "completed": [ 256, 256, 256 ], "ttft_quantiles_ms": [ 157.6, 2168.2, 2201.8, 3880.2, 5051.1, 5845.1, 7671.6, 8232.3, 10737.3, 10927.0, 11015.6, 11130.5, 11477.1, 11530.7, 12422.5, 12776.5, 12837.8, 13064.7, 13163.7, 21645.0, 24269.8, 24749.1, 26838.1, 27023.6, 27103.3, 27258.9, 27413.6, 27543.6, 27580.9, 28020.4, 28350.7, 29511.0, 29732.2, 29768.4, 30093.4, 30285.7, 30318.8, 30537.6, 30712.3, 31111.6, 32296.4, 32421.4, 32782.6, 32976.3, 33077.0, 33316.9, 33357.2, 33697.7, 35113.9, 35799.9, 36373.0 ], "tpot_quantiles_ms": [ 84.7, 85.6, 88.2, 89.5, 90.4, 92.8, 93.0, 93.8, 95.7, 98.6, 100.2, 102.5, 102.8, 104.1, 104.6, 106.7, 107.5, 109.5, 110.0, 113.6, 118.1, 120.4, 121.4, 122.2, 123.6, 123.9, 125.1, 127.9, 129.1, 131.3, 133.6, 135.1, 136.6, 142.2, 143.0, 144.0, 144.8, 146.2, 147.4, 149.0, 150.9, 153.6, 162.4, 164.0, 164.3, 165.9, 167.2, 167.8, 170.1, 171.4, 173.3 ] } ] }, "eagle3": { "kv_pool_tokens": 9951, "points": [ { "x": 1.0, "output_throughput": { "median": 29.7, "min": 26.58, "max": 30.17, "n": 3 }, "median_ttft_ms": { "median": 476.73, "min": 265.03, "max": 587.7, "n": 3 }, "p99_ttft_ms": { "median": 709.44, "min": 553.34, "max": 729.6, "n": 3 }, "median_tpot_ms": { "median": 25.66, "min": 25.32, "max": 30.92, "n": 3 }, "p99_tpot_ms": { "median": 61.96, "min": 51.05, "max": 80.09, "n": 3 }, "goodput_rps": { "median": 0.23, "min": 0.21, "max": 0.24, "n": 3 }, "request_throughput": { "median": 0.23, "min": 0.21, "max": 0.24, "n": 3 }, "completed": [ 16, 16, 16 ], "accept_length": { "median": 2.72, "min": 2.5, "max": 2.74, "n": 3 }, "ttft_quantiles_ms": [ 85.5, 85.5, 86.7, 88.3, 202.7, 203.2, 203.4, 214.1, 228.0, 257.9, 262.9, 264.2, 265.2, 267.1, 290.5, 292.3, 292.9, 292.9, 298.3, 322.1, 343.7, 349.4, 378.9, 387.5, 403.4, 432.7, 468.4, 480.7, 485.1, 514.7, 520.7, 529.8, 531.4, 541.9, 541.9, 544.9, 548.7, 554.8, 562.2, 611.1, 613.2, 647.1, 660.9, 668.4, 670.1, 680.7, 688.4, 691.9, 712.3, 712.5, 732.7 ], "tpot_quantiles_ms": [ 18.5, 18.5, 20.2, 20.3, 20.7, 20.8, 20.8, 21.0, 21.3, 21.3, 22.1, 22.2, 22.6, 23.6, 23.6, 23.7, 24.4, 24.4, 25.3, 25.3, 25.3, 25.4, 25.9, 26.1, 27.1, 27.6, 27.7, 28.8, 30.5, 30.6, 30.7, 30.8, 31.0, 31.3, 31.3, 32.4, 33.3, 33.8, 34.8, 35.2, 36.3, 38.5, 39.8, 40.3, 44.3, 47.5, 50.1, 51.2, 56.7, 62.9, 85.8 ] }, { "x": 2.0, "output_throughput": { "median": 47.85, "min": 46.95, "max": 56.01, "n": 3 }, "median_ttft_ms": { "median": 506.85, "min": 473.09, "max": 589.51, "n": 3 }, "p99_ttft_ms": { "median": 773.56, "min": 651.71, "max": 781.13, "n": 3 }, "median_tpot_ms": { "median": 30.6, "min": 30.13, "max": 30.63, "n": 3 }, "p99_tpot_ms": { "median": 66.57, "min": 47.81, "max": 71.08, "n": 3 }, "goodput_rps": { "median": 0.37, "min": 0.37, "max": 0.44, "n": 3 }, "request_throughput": { "median": 0.37, "min": 0.37, "max": 0.44, "n": 3 }, "completed": [ 16, 16, 16 ], "accept_length": { "median": 2.72, "min": 2.49, "max": 2.74, "n": 3 }, "ttft_quantiles_ms": [ 238.8, 238.8, 247.9, 248.7, 250.8, 263.0, 265.2, 329.9, 332.4, 333.2, 356.9, 370.0, 373.7, 386.0, 390.7, 391.9, 392.1, 392.1, 452.3, 455.1, 471.2, 474.5, 475.0, 479.5, 500.9, 512.8, 518.8, 519.6, 541.8, 551.9, 562.6, 580.9, 588.2, 598.1, 598.1, 603.7, 604.5, 606.3, 612.1, 643.5, 649.2, 649.7, 652.1, 678.5, 681.7, 685.9, 704.2, 751.3, 766.6, 774.8, 797.9 ], "tpot_quantiles_ms": [ 18.9, 18.9, 20.6, 21.2, 21.3, 21.7, 21.9, 22.5, 23.2, 24.5, 24.9, 25.0, 25.9, 26.8, 26.9, 27.2, 27.2, 27.2, 28.4, 28.5, 28.7, 29.5, 29.6, 29.8, 30.2, 30.4, 30.6, 30.9, 30.9, 31.3, 31.9, 32.8, 33.0, 33.0, 33.0, 33.0, 33.2, 33.6, 35.7, 36.5, 39.3, 42.3, 42.6, 42.9, 44.8, 44.9, 46.9, 48.0, 64.1, 67.0, 75.7 ] }, { "x": 4.0, "output_throughput": { "median": 87.38, "min": 83.82, "max": 111.15, "n": 3 }, "median_ttft_ms": { "median": 297.63, "min": 291.25, "max": 336.3, "n": 3 }, "p99_ttft_ms": { "median": 590.41, "min": 479.14, "max": 777.43, "n": 3 }, "median_tpot_ms": { "median": 31.23, "min": 29.08, "max": 34.77, "n": 3 }, "p99_tpot_ms": { "median": 55.95, "min": 48.74, "max": 63.77, "n": 3 }, "goodput_rps": { "median": 0.68, "min": 0.65, "max": 0.87, "n": 3 }, "request_throughput": { "median": 0.68, "min": 0.65, "max": 0.87, "n": 3 }, "completed": [ 16, 16, 16 ], "accept_length": { "median": 2.72, "min": 2.48, "max": 2.73, "n": 3 }, "ttft_quantiles_ms": [ 154.7, 154.7, 218.5, 218.7, 218.9, 219.1, 219.6, 220.4, 220.5, 220.8, 220.9, 221.1, 221.2, 222.1, 222.2, 222.5, 224.8, 224.8, 265.1, 274.9, 276.0, 286.0, 289.9, 290.5, 291.0, 291.5, 305.4, 306.1, 306.4, 310.8, 332.7, 334.3, 338.3, 351.0, 351.0, 352.2, 358.0, 358.6, 358.9, 360.1, 369.5, 398.3, 433.8, 444.3, 487.1, 507.1, 523.7, 616.2, 642.5, 776.5, 777.6 ], "tpot_quantiles_ms": [ 19.4, 19.4, 21.1, 21.2, 21.2, 21.8, 22.4, 22.5, 22.9, 23.0, 23.7, 24.8, 24.9, 24.9, 25.5, 25.6, 25.8, 25.8, 26.2, 27.2, 28.6, 29.0, 29.2, 29.9, 30.7, 30.8, 31.6, 32.0, 34.1, 34.5, 34.7, 35.1, 35.7, 37.1, 37.1, 37.2, 37.4, 39.1, 39.2, 40.6, 40.9, 40.9, 42.6, 44.1, 44.6, 45.4, 49.3, 51.1, 56.8, 61.0, 64.3 ] }, { "x": 8.0, "output_throughput": { "median": 157.08, "min": 137.43, "max": 159.65, "n": 3 }, "median_ttft_ms": { "median": 426.38, "min": 410.25, "max": 435.83, "n": 3 }, "p99_ttft_ms": { "median": 1033.64, "min": 973.13, "max": 1102.21, "n": 3 }, "median_tpot_ms": { "median": 42.3, "min": 37.73, "max": 44.14, "n": 3 }, "p99_tpot_ms": { "median": 107.54, "min": 64.76, "max": 122.38, "n": 3 }, "goodput_rps": { "median": 1.01, "min": 1.01, "max": 1.23, "n": 3 }, "request_throughput": { "median": 1.23, "min": 1.07, "max": 1.25, "n": 3 }, "completed": [ 32, 32, 32 ], "accept_length": { "median": 2.72, "min": 2.59, "max": 2.73, "n": 3 }, "ttft_quantiles_ms": [ 218.0, 219.6, 220.5, 234.2, 235.1, 235.5, 236.8, 237.5, 245.2, 292.5, 293.6, 293.8, 299.0, 299.6, 300.3, 309.8, 310.5, 311.2, 311.8, 334.9, 335.5, 335.8, 336.5, 376.5, 393.5, 426.6, 448.2, 459.3, 467.1, 474.3, 492.2, 495.2, 516.7, 543.1, 577.7, 581.0, 600.5, 625.1, 632.6, 655.0, 665.4, 724.2, 845.1, 889.2, 907.8, 954.6, 975.8, 980.8, 1051.8, 1057.4, 1102.4 ], "tpot_quantiles_ms": [ 21.8, 22.9, 24.3, 24.4, 25.1, 25.2, 25.9, 27.4, 28.7, 29.3, 30.3, 30.3, 31.7, 32.2, 33.1, 33.8, 34.7, 35.2, 35.4, 37.3, 37.7, 37.8, 38.4, 39.1, 39.7, 40.5, 40.7, 41.9, 42.7, 43.5, 45.3, 45.3, 45.9, 47.0, 47.6, 49.0, 50.6, 50.8, 51.2, 52.9, 55.6, 55.8, 56.4, 57.0, 57.4, 63.7, 64.4, 64.9, 73.6, 90.9, 137.2 ] }, { "x": 16.0, "output_throughput": { "median": 191.76, "min": 184.98, "max": 200.88, "n": 3 }, "median_ttft_ms": { "median": 1469.45, "min": 1441.05, "max": 1576.24, "n": 3 }, "p99_ttft_ms": { "median": 4442.37, "min": 4131.25, "max": 4518.62, "n": 3 }, "median_tpot_ms": { "median": 60.61, "min": 56.9, "max": 61.78, "n": 3 }, "p99_tpot_ms": { "median": 135.7, "min": 113.0, "max": 146.9, "n": 3 }, "goodput_rps": { "median": 0.3, "min": 0.16, "max": 0.32, "n": 3 }, "request_throughput": { "median": 1.5, "min": 1.45, "max": 1.57, "n": 3 }, "completed": [ 64, 64, 64 ], "accept_length": { "median": 2.71, "min": 2.65, "max": 2.73, "n": 3 }, "ttft_quantiles_ms": [ 523.8, 674.0, 777.2, 818.9, 834.3, 861.3, 876.2, 915.5, 975.4, 986.6, 1030.3, 1051.8, 1073.2, 1091.6, 1115.4, 1138.2, 1171.5, 1183.8, 1206.9, 1231.9, 1293.1, 1330.7, 1400.1, 1411.8, 1465.1, 1487.8, 1528.1, 1579.5, 1602.5, 1633.9, 1716.8, 1759.2, 1782.7, 1877.5, 1943.3, 2114.0, 2248.0, 2374.5, 2620.7, 2825.4, 2825.8, 2880.8, 2881.1, 2904.4, 2906.1, 2912.7, 2914.3, 2955.7, 2958.8, 2960.0, 7306.9 ], "tpot_quantiles_ms": [ 24.2, 26.0, 28.4, 30.2, 33.3, 36.2, 38.2, 40.2, 42.3, 44.5, 45.7, 46.7, 47.7, 48.5, 49.1, 50.8, 52.1, 53.0, 53.9, 55.1, 55.7, 56.3, 57.1, 58.0, 59.0, 59.3, 60.1, 60.6, 60.7, 61.7, 62.8, 64.7, 65.9, 66.7, 68.5, 71.1, 73.5, 75.1, 77.1, 78.5, 80.8, 83.5, 85.0, 91.3, 94.5, 97.7, 102.4, 106.9, 112.5, 120.6, 154.4 ] }, { "x": 32.0, "output_throughput": { "median": 192.04, "min": 177.74, "max": 192.04, "n": 3 }, "median_ttft_ms": { "median": 12088.31, "min": 12065.02, "max": 12894.03, "n": 3 }, "p99_ttft_ms": { "median": 14803.17, "min": 14257.05, "max": 15596.93, "n": 3 }, "median_tpot_ms": { "median": 64.37, "min": 61.44, "max": 68.83, "n": 3 }, "p99_tpot_ms": { "median": 148.66, "min": 127.71, "max": 149.22, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.5, "min": 1.39, "max": 1.5, "n": 3 }, "completed": [ 128, 128, 128 ], "accept_length": { "median": 2.72, "min": 2.71, "max": 2.73, "n": 3 }, "ttft_quantiles_ms": [ 1389.8, 2699.4, 2747.2, 2801.6, 2856.5, 3177.8, 6437.3, 7009.8, 8419.1, 9193.0, 9565.9, 9873.1, 10393.1, 10758.3, 11242.6, 11466.0, 11577.3, 11654.5, 11777.4, 11830.9, 11935.7, 12057.2, 12118.3, 12238.6, 12292.0, 12328.7, 12407.7, 12449.7, 12483.3, 12522.7, 12571.0, 12587.2, 12630.5, 12682.1, 12770.2, 12851.0, 12895.1, 13017.7, 13054.6, 13158.7, 13250.3, 13312.6, 13384.2, 13535.4, 13622.2, 13678.0, 13835.0, 14044.1, 14257.1, 14766.8, 16444.3 ], "tpot_quantiles_ms": [ 24.2, 28.5, 35.0, 38.7, 42.6, 44.8, 45.9, 48.2, 49.8, 50.8, 52.1, 53.2, 54.1, 54.8, 56.7, 57.5, 58.0, 58.9, 59.4, 60.0, 61.2, 61.7, 62.4, 63.0, 63.8, 64.4, 64.9, 65.6, 66.8, 68.6, 69.4, 70.5, 73.4, 74.1, 76.7, 78.8, 80.7, 82.5, 85.2, 88.0, 90.8, 92.6, 95.6, 99.4, 101.9, 103.4, 109.2, 116.5, 123.6, 128.8, 172.9 ] }, { "x": 64.0, "output_throughput": { "median": 180.18, "min": 167.19, "max": 188.06, "n": 3 }, "median_ttft_ms": { "median": 35563.96, "min": 34464.63, "max": 37402.7, "n": 3 }, "p99_ttft_ms": { "median": 39782.51, "min": 36707.73, "max": 42682.27, "n": 3 }, "median_tpot_ms": { "median": 69.7, "min": 67.81, "max": 75.26, "n": 3 }, "p99_tpot_ms": { "median": 152.16, "min": 137.0, "max": 162.52, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.41, "min": 1.31, "max": 1.47, "n": 3 }, "completed": [ 256, 256, 256 ], "accept_length": { "median": 2.73, "min": 2.71, "max": 2.75, "n": 3 }, "ttft_quantiles_ms": [ 1464.8, 2648.2, 2890.3, 6484.4, 8654.2, 11354.9, 14541.1, 17742.9, 21658.8, 25957.2, 29086.0, 31729.2, 32921.8, 33453.9, 33728.2, 33977.0, 34149.5, 34321.0, 34466.5, 34626.2, 34737.2, 34915.3, 35030.1, 35127.5, 35299.7, 35399.3, 35535.1, 35636.8, 35723.1, 35816.5, 35908.5, 36062.0, 36234.6, 36382.6, 36671.2, 36778.5, 37052.4, 37302.3, 37780.5, 38031.1, 38201.3, 38480.1, 38773.0, 38973.2, 39224.2, 39506.0, 40003.2, 40384.9, 40873.4, 41277.9, 42946.6 ], "tpot_quantiles_ms": [ 24.2, 33.2, 40.8, 47.0, 49.7, 51.4, 52.7, 54.0, 55.5, 56.8, 58.0, 58.5, 59.0, 59.7, 60.8, 61.5, 62.2, 62.7, 63.9, 64.7, 65.4, 66.6, 67.3, 68.3, 69.8, 70.5, 71.3, 72.2, 73.2, 74.9, 77.2, 78.7, 80.0, 81.5, 84.0, 85.8, 88.0, 90.7, 93.6, 96.3, 99.4, 101.7, 103.7, 106.2, 108.7, 112.7, 115.7, 122.3, 128.2, 139.4, 227.2 ] } ] }, "fp8": { "kv_pool_tokens": 46228, "points": [ { "x": 1.0, "output_throughput": { "median": 7.93, "min": 7.84, "max": 7.94, "n": 3 }, "median_ttft_ms": { "median": 689.74, "min": 674.59, "max": 711.99, "n": 3 }, "p99_ttft_ms": { "median": 944.05, "min": 754.25, "max": 980.87, "n": 3 }, "median_tpot_ms": { "median": 124.39, "min": 124.09, "max": 127.02, "n": 3 }, "p99_tpot_ms": { "median": 137.22, "min": 128.77, "max": 144.26, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 0.06, "min": 0.06, "max": 0.06, "n": 3 }, "completed": [ 16, 16, 16 ], "ttft_quantiles_ms": [ 61.3, 61.3, 85.3, 151.4, 270.2, 366.5, 408.9, 448.4, 480.2, 494.9, 519.7, 569.8, 615.1, 622.1, 625.4, 645.0, 645.1, 645.1, 655.5, 657.2, 669.6, 670.2, 678.9, 681.9, 682.1, 685.3, 688.5, 688.9, 689.1, 691.0, 695.3, 696.9, 708.3, 719.7, 719.7, 722.7, 725.3, 728.7, 734.1, 746.9, 752.6, 759.3, 777.1, 778.6, 800.9, 837.9, 854.0, 866.3, 881.8, 955.0, 1001.1 ], "tpot_quantiles_ms": [ 72.3, 72.3, 77.8, 84.3, 109.1, 110.5, 111.7, 114.9, 115.8, 117.0, 117.2, 118.4, 118.9, 119.7, 121.0, 121.5, 121.7, 121.7, 122.5, 123.6, 123.9, 124.3, 125.1, 125.4, 126.0, 126.4, 126.8, 126.9, 127.0, 127.0, 127.0, 127.3, 127.4, 127.9, 127.9, 127.9, 128.0, 128.1, 128.4, 128.7, 128.8, 129.2, 129.6, 129.8, 130.8, 131.4, 133.9, 135.2, 135.8, 137.5, 146.5 ] }, { "x": 2.0, "output_throughput": { "median": 15.99, "min": 15.69, "max": 16.29, "n": 3 }, "median_ttft_ms": { "median": 232.78, "min": 211.97, "max": 258.16, "n": 3 }, "p99_ttft_ms": { "median": 368.88, "min": 337.04, "max": 374.77, "n": 3 }, "median_tpot_ms": { "median": 127.63, "min": 126.88, "max": 131.37, "n": 3 }, "p99_tpot_ms": { "median": 145.1, "min": 143.22, "max": 149.94, "n": 3 }, "goodput_rps": { "median": 0.02, "min": 0.0, "max": 0.02, "n": 3 }, "request_throughput": { "median": 0.12, "min": 0.12, "max": 0.13, "n": 3 }, "completed": [ 16, 16, 16 ], "ttft_quantiles_ms": [ 104.9, 104.9, 105.8, 109.8, 112.4, 115.3, 124.8, 125.3, 150.7, 165.4, 168.5, 170.1, 177.6, 180.4, 183.8, 185.2, 189.7, 189.7, 197.6, 205.4, 207.8, 209.6, 209.8, 213.5, 226.4, 229.1, 240.8, 252.1, 254.2, 255.8, 257.5, 260.5, 265.8, 269.7, 269.7, 271.7, 276.6, 278.0, 289.9, 302.7, 303.0, 303.9, 316.6, 319.8, 325.7, 334.6, 337.5, 347.3, 362.4, 377.0, 377.5 ], "tpot_quantiles_ms": [ 80.4, 80.4, 81.0, 87.6, 87.7, 102.2, 102.2, 103.8, 104.6, 104.9, 105.5, 112.2, 112.8, 118.1, 118.5, 119.0, 119.7, 119.7, 122.9, 123.9, 124.6, 125.7, 126.5, 126.6, 127.4, 127.8, 128.0, 128.1, 128.7, 129.0, 129.0, 129.1, 132.1, 133.2, 133.2, 134.9, 135.7, 135.8, 136.9, 137.5, 138.2, 138.4, 139.1, 142.2, 143.3, 143.4, 144.5, 144.7, 145.2, 149.2, 150.1 ] }, { "x": 4.0, "output_throughput": { "median": 35.04, "min": 33.09, "max": 36.0, "n": 3 }, "median_ttft_ms": { "median": 224.46, "min": 118.95, "max": 295.01, "n": 3 }, "p99_ttft_ms": { "median": 373.3, "min": 289.77, "max": 380.66, "n": 3 }, "median_tpot_ms": { "median": 117.57, "min": 114.56, "max": 123.24, "n": 3 }, "p99_tpot_ms": { "median": 129.96, "min": 124.42, "max": 139.36, "n": 3 }, "goodput_rps": { "median": 0.07, "min": 0.06, "max": 0.07, "n": 3 }, "request_throughput": { "median": 0.27, "min": 0.26, "max": 0.28, "n": 3 }, "completed": [ 16, 16, 16 ], "ttft_quantiles_ms": [ 76.6, 76.6, 77.3, 78.0, 101.6, 101.8, 102.2, 102.4, 102.5, 102.7, 105.0, 118.0, 118.7, 119.2, 130.0, 130.6, 131.3, 131.3, 164.4, 166.9, 193.6, 199.6, 199.8, 199.8, 214.7, 215.3, 215.7, 220.6, 224.1, 224.4, 224.5, 225.0, 283.8, 284.3, 284.3, 284.7, 288.7, 289.2, 289.9, 294.8, 295.3, 295.7, 302.5, 303.0, 303.6, 372.3, 372.9, 373.4, 379.6, 380.1, 380.8 ], "tpot_quantiles_ms": [ 88.3, 88.3, 88.3, 88.3, 88.5, 89.9, 89.9, 89.9, 90.1, 94.2, 94.2, 94.2, 94.6, 111.2, 111.2, 111.2, 111.9, 111.9, 112.8, 112.8, 112.9, 113.4, 115.7, 115.7, 115.7, 115.9, 119.9, 119.9, 119.9, 120.3, 123.2, 123.2, 123.2, 124.1, 124.1, 124.2, 124.2, 124.3, 124.5, 126.1, 126.1, 126.1, 126.4, 128.7, 128.7, 128.7, 130.2, 138.2, 138.2, 138.2, 139.6 ] }, { "x": 8.0, "output_throughput": { "median": 57.86, "min": 57.61, "max": 58.65, "n": 3 }, "median_ttft_ms": { "median": 424.55, "min": 422.71, "max": 446.1, "n": 3 }, "p99_ttft_ms": { "median": 6180.01, "min": 6078.51, "max": 6666.81, "n": 3 }, "median_tpot_ms": { "median": 120.39, "min": 117.44, "max": 123.04, "n": 3 }, "p99_tpot_ms": { "median": 164.62, "min": 158.56, "max": 169.29, "n": 3 }, "goodput_rps": { "median": 0.11, "min": 0.0, "max": 0.11, "n": 3 }, "request_throughput": { "median": 0.45, "min": 0.45, "max": 0.46, "n": 3 }, "completed": [ 32, 32, 32 ], "ttft_quantiles_ms": [ 79.9, 84.9, 97.1, 105.4, 106.1, 106.7, 110.7, 111.3, 112.0, 112.6, 181.5, 182.1, 183.1, 184.5, 186.9, 206.4, 207.7, 209.0, 245.9, 246.5, 247.1, 247.5, 339.1, 340.2, 341.6, 343.0, 639.3, 730.1, 808.9, 3466.9, 3468.4, 3469.3, 4637.3, 4638.0, 4638.8, 4661.7, 4662.9, 5362.4, 5363.1, 5364.1, 5365.2, 5586.9, 5587.9, 6078.7, 6176.9, 6178.1, 6179.6, 6662.5, 6663.9, 6665.5, 6667.0 ], "tpot_quantiles_ms": [ 91.7, 91.7, 91.7, 91.7, 92.2, 99.5, 99.5, 99.5, 99.7, 103.0, 103.0, 103.1, 103.1, 115.7, 115.7, 115.8, 115.8, 117.1, 117.1, 117.1, 117.1, 117.4, 117.4, 117.5, 117.5, 117.5, 117.6, 123.0, 123.0, 123.0, 123.1, 124.2, 124.2, 124.3, 124.3, 128.2, 128.2, 128.2, 128.2, 128.8, 128.8, 128.8, 129.7, 129.7, 129.7, 134.1, 134.1, 145.7, 156.4, 166.9, 171.1 ] }, { "x": 16.0, "output_throughput": { "median": 98.74, "min": 96.2, "max": 100.24, "n": 3 }, "median_ttft_ms": { "median": 598.16, "min": 512.29, "max": 605.54, "n": 3 }, "p99_ttft_ms": { "median": 11975.09, "min": 11205.74, "max": 12327.78, "n": 3 }, "median_tpot_ms": { "median": 134.76, "min": 132.4, "max": 142.61, "n": 3 }, "p99_tpot_ms": { "median": 197.74, "min": 193.37, "max": 198.47, "n": 3 }, "goodput_rps": { "median": 0.19, "min": 0.0, "max": 0.2, "n": 3 }, "request_throughput": { "median": 0.77, "min": 0.75, "max": 0.78, "n": 3 }, "completed": [ 64, 64, 64 ], "ttft_quantiles_ms": [ 84.2, 125.3, 125.8, 126.5, 127.1, 127.8, 128.3, 129.0, 146.7, 240.3, 246.6, 247.9, 248.5, 250.0, 324.5, 326.3, 328.4, 399.4, 401.0, 402.3, 403.3, 404.2, 405.7, 407.2, 409.9, 477.7, 765.5, 2551.4, 3256.5, 3735.1, 3859.9, 3864.9, 4169.4, 5635.9, 5641.6, 7317.0, 7318.9, 8640.1, 8643.4, 9566.9, 9777.6, 9780.7, 9833.8, 9836.3, 10254.8, 10256.5, 10258.0, 11205.9, 11932.4, 11934.2, 12998.0 ], "tpot_quantiles_ms": [ 97.2, 97.7, 97.7, 97.7, 98.4, 98.5, 98.5, 98.6, 98.9, 106.2, 111.0, 111.3, 111.4, 111.4, 113.5, 117.2, 117.2, 117.3, 121.9, 121.9, 123.5, 123.5, 131.7, 132.4, 132.4, 134.3, 134.8, 134.8, 135.0, 135.0, 135.0, 139.8, 139.8, 139.8, 139.8, 141.1, 142.6, 142.6, 143.4, 143.4, 143.4, 143.4, 167.6, 169.9, 171.1, 171.5, 171.5, 179.4, 185.0, 192.1, 209.4 ] }, { "x": 32.0, "output_throughput": { "median": 154.99, "min": 154.31, "max": 156.73, "n": 3 }, "median_ttft_ms": { "median": 517.39, "min": 409.48, "max": 567.61, "n": 3 }, "p99_ttft_ms": { "median": 19625.23, "min": 18910.25, "max": 20978.92, "n": 3 }, "median_tpot_ms": { "median": 156.5, "min": 127.73, "max": 161.63, "n": 3 }, "p99_tpot_ms": { "median": 272.12, "min": 258.25, "max": 280.96, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.21, "min": 1.21, "max": 1.22, "n": 3 }, "completed": [ 128, 128, 128 ], "ttft_quantiles_ms": [ 111.9, 176.2, 181.3, 183.8, 199.6, 202.3, 209.0, 211.5, 224.8, 226.6, 229.2, 318.5, 322.5, 333.9, 337.3, 340.2, 349.3, 357.0, 362.4, 370.1, 374.2, 378.1, 401.1, 402.9, 404.6, 408.0, 1413.9, 4315.4, 4509.8, 4663.6, 6024.3, 7953.4, 8576.0, 8828.2, 9278.7, 9977.7, 10621.6, 12908.8, 13648.4, 14385.0, 14504.5, 15302.3, 15309.5, 15381.1, 16950.2, 18142.8, 18286.8, 19140.0, 19711.3, 20516.4, 21180.3 ], "tpot_quantiles_ms": [ 107.7, 107.7, 107.8, 107.9, 107.9, 108.9, 108.9, 110.6, 110.6, 111.8, 111.8, 111.9, 111.9, 127.1, 127.1, 127.5, 127.5, 127.7, 133.5, 134.1, 136.4, 137.7, 138.7, 139.9, 156.4, 156.5, 156.5, 156.6, 156.6, 161.6, 161.6, 161.7, 161.8, 161.9, 168.2, 170.7, 174.0, 174.5, 178.7, 193.9, 200.4, 216.4, 216.4, 218.9, 219.9, 230.2, 235.9, 247.6, 258.3, 261.7, 289.9 ] }, { "x": 64.0, "output_throughput": { "median": 155.06, "min": 154.48, "max": 156.28, "n": 3 }, "median_ttft_ms": { "median": 19862.63, "min": 19829.23, "max": 20109.71, "n": 3 }, "p99_ttft_ms": { "median": 37460.16, "min": 37439.37, "max": 38109.46, "n": 3 }, "median_tpot_ms": { "median": 253.45, "min": 253.09, "max": 258.01, "n": 3 }, "p99_tpot_ms": { "median": 413.91, "min": 401.2, "max": 418.5, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.21, "min": 1.21, "max": 1.22, "n": 3 }, "completed": [ 256, 256, 256 ], "ttft_quantiles_ms": [ 194.7, 2639.5, 2989.1, 3563.2, 4476.1, 4636.2, 5718.9, 6454.0, 8091.3, 8708.9, 9482.1, 9912.5, 10096.6, 10835.1, 12795.6, 13120.8, 13342.0, 13646.1, 14847.2, 15630.9, 15986.0, 16763.8, 17701.8, 18532.7, 19650.2, 20002.7, 20125.1, 20416.0, 21351.1, 22688.8, 23501.0, 23521.7, 24246.7, 25001.0, 25416.5, 26907.3, 27014.1, 27558.7, 28421.5, 29448.8, 30199.6, 30504.9, 31740.8, 33206.8, 33487.5, 34109.9, 35625.0, 35921.7, 36453.2, 37459.2, 38308.5 ], "tpot_quantiles_ms": [ 134.9, 137.7, 142.7, 144.6, 146.5, 150.1, 156.3, 157.8, 167.6, 172.9, 177.9, 178.3, 185.4, 193.7, 199.2, 203.5, 213.9, 216.3, 221.1, 221.9, 226.2, 234.9, 242.0, 249.7, 252.9, 253.4, 257.3, 274.3, 276.5, 280.6, 281.3, 285.7, 285.8, 306.9, 307.3, 312.4, 315.9, 321.4, 324.3, 332.0, 334.6, 343.1, 351.3, 356.4, 361.2, 368.9, 378.3, 396.7, 401.6, 408.1, 441.7 ] } ] }, "no-cuda-graph": { "kv_pool_tokens": 24094, "points": [ { "x": 1.0, "output_throughput": { "median": 10.24, "min": 10.18, "max": 12.26, "n": 3 }, "median_ttft_ms": { "median": 713.05, "min": 462.66, "max": 716.27, "n": 3 }, "p99_ttft_ms": { "median": 723.63, "min": 714.45, "max": 727.49, "n": 3 }, "median_tpot_ms": { "median": 103.41, "min": 84.66, "max": 103.41, "n": 3 }, "p99_tpot_ms": { "median": 103.47, "min": 103.46, "max": 103.48, "n": 3 }, "goodput_rps": { "median": 0.03, "min": 0.03, "max": 0.07, "n": 3 }, "request_throughput": { "median": 0.08, "min": 0.08, "max": 0.1, "n": 3 }, "completed": [ 16, 16, 16 ], "ttft_quantiles_ms": [ 90.5, 90.5, 91.7, 92.9, 194.9, 199.2, 200.7, 203.7, 208.2, 220.7, 302.5, 312.8, 435.1, 458.3, 463.5, 465.6, 490.2, 490.2, 494.4, 495.0, 498.2, 499.6, 502.1, 503.8, 506.4, 652.8, 712.2, 712.6, 712.7, 713.4, 713.5, 713.5, 713.9, 713.9, 713.9, 714.3, 714.5, 715.2, 716.2, 716.4, 716.6, 717.2, 721.0, 721.2, 721.5, 721.7, 723.5, 723.6, 723.6, 724.8, 728.0 ], "tpot_quantiles_ms": [ 60.6, 60.6, 60.6, 60.6, 60.6, 60.6, 60.6, 60.7, 60.7, 61.1, 64.4, 66.5, 75.4, 75.8, 81.3, 84.3, 85.0, 85.0, 85.1, 85.3, 85.8, 85.8, 86.0, 86.4, 90.2, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.4, 103.5, 103.5, 103.5, 103.5, 103.5, 103.5 ] }, { "x": 2.0, "output_throughput": { "median": 26.62, "min": 25.62, "max": 27.33, "n": 3 }, "median_ttft_ms": { "median": 141.11, "min": 129.03, "max": 149.13, "n": 3 }, "p99_ttft_ms": { "median": 226.22, "min": 191.94, "max": 236.16, "n": 3 }, "median_tpot_ms": { "median": 73.9, "min": 71.94, "max": 76.31, "n": 3 }, "p99_tpot_ms": { "median": 92.88, "min": 76.9, "max": 100.48, "n": 3 }, "goodput_rps": { "median": 0.2, "min": 0.19, "max": 0.21, "n": 3 }, "request_throughput": { "median": 0.21, "min": 0.2, "max": 0.21, "n": 3 }, "completed": [ 16, 16, 16 ], "ttft_quantiles_ms": [ 97.4, 97.4, 98.0, 98.1, 98.2, 98.8, 98.8, 99.2, 99.5, 100.4, 100.8, 101.0, 101.2, 101.4, 101.5, 101.9, 102.1, 102.1, 102.2, 103.6, 103.6, 104.1, 104.1, 113.9, 127.3, 143.5, 154.0, 154.8, 154.9, 156.3, 156.3, 156.6, 156.6, 158.4, 158.4, 159.8, 159.9, 164.8, 170.6, 174.5, 180.1, 183.8, 184.3, 185.5, 186.5, 188.0, 188.7, 192.5, 214.4, 228.3, 244.9 ], "tpot_quantiles_ms": [ 64.7, 64.7, 64.9, 65.2, 65.3, 65.3, 65.4, 65.7, 65.8, 66.1, 66.5, 71.3, 71.4, 71.4, 71.5, 71.5, 71.6, 71.6, 71.9, 72.0, 72.0, 72.0, 72.1, 72.1, 72.1, 72.2, 72.7, 72.7, 72.8, 72.8, 73.1, 73.2, 74.6, 75.2, 75.2, 75.7, 75.8, 76.3, 76.3, 76.5, 77.0, 79.5, 80.0, 85.2, 86.0, 86.8, 87.6, 92.5, 92.9, 99.8, 100.6 ] }, { "x": 4.0, "output_throughput": { "median": 55.49, "min": 55.35, "max": 58.68, "n": 3 }, "median_ttft_ms": { "median": 161.64, "min": 160.14, "max": 162.22, "n": 3 }, "p99_ttft_ms": { "median": 241.11, "min": 165.99, "max": 246.82, "n": 3 }, "median_tpot_ms": { "median": 68.0, "min": 67.92, "max": 69.75, "n": 3 }, "p99_tpot_ms": { "median": 81.67, "min": 69.28, "max": 84.44, "n": 3 }, "goodput_rps": { "median": 0.43, "min": 0.43, "max": 0.46, "n": 3 }, "request_throughput": { "median": 0.43, "min": 0.43, "max": 0.46, "n": 3 }, "completed": [ 16, 16, 16 ], "ttft_quantiles_ms": [ 98.2, 98.2, 100.0, 100.2, 100.2, 101.1, 101.2, 101.6, 101.6, 102.7, 103.0, 134.4, 140.5, 156.8, 157.4, 157.7, 159.3, 159.3, 159.8, 159.9, 159.9, 160.2, 160.3, 160.4, 160.7, 160.8, 161.4, 161.9, 162.0, 162.0, 162.1, 162.4, 162.4, 162.5, 162.5, 162.5, 162.6, 162.9, 162.9, 163.0, 163.3, 163.6, 165.0, 165.5, 166.1, 240.4, 240.7, 241.2, 245.9, 246.4, 246.9 ], "tpot_quantiles_ms": [ 65.3, 65.3, 65.3, 65.3, 65.3, 65.3, 65.3, 65.4, 65.4, 65.4, 65.7, 65.8, 65.9, 66.5, 66.5, 66.5, 66.5, 66.5, 66.5, 66.5, 66.7, 66.7, 66.7, 66.9, 67.0, 67.2, 68.6, 68.6, 68.6, 68.8, 68.8, 68.8, 69.1, 69.1, 69.1, 69.1, 69.1, 69.3, 69.6, 72.5, 72.5, 72.5, 73.0, 81.0, 81.0, 81.0, 81.8, 83.7, 83.7, 83.7, 84.6 ] }, { "x": 8.0, "output_throughput": { "median": 92.0, "min": 91.18, "max": 92.76, "n": 3 }, "median_ttft_ms": { "median": 312.08, "min": 307.5, "max": 364.72, "n": 3 }, "p99_ttft_ms": { "median": 3892.12, "min": 3648.46, "max": 4193.93, "n": 3 }, "median_tpot_ms": { "median": 75.36, "min": 72.62, "max": 77.98, "n": 3 }, "p99_tpot_ms": { "median": 99.57, "min": 97.77, "max": 102.8, "n": 3 }, "goodput_rps": { "median": 0.4, "min": 0.38, "max": 0.41, "n": 3 }, "request_throughput": { "median": 0.72, "min": 0.71, "max": 0.72, "n": 3 }, "completed": [ 32, 32, 32 ], "ttft_quantiles_ms": [ 101.2, 102.2, 103.6, 120.2, 161.1, 162.3, 162.8, 163.2, 163.4, 164.0, 164.3, 164.4, 165.1, 165.5, 166.0, 166.4, 166.9, 167.6, 168.5, 169.2, 179.2, 179.5, 180.1, 180.8, 220.9, 226.5, 443.0, 562.1, 579.9, 2451.1, 2452.6, 2453.1, 2750.8, 2751.8, 2753.2, 2801.5, 2898.1, 2899.5, 2900.9, 2917.0, 2918.3, 2919.0, 2960.4, 2961.9, 3255.8, 3648.1, 3891.6, 4190.5, 4191.5, 4192.9, 4194.1 ], "tpot_quantiles_ms": [ 66.8, 66.8, 66.8, 66.8, 66.9, 66.9, 66.9, 66.9, 67.0, 67.4, 67.5, 67.5, 67.5, 70.7, 71.3, 71.3, 71.3, 71.8, 73.5, 73.5, 73.6, 73.6, 74.3, 75.4, 75.4, 75.4, 75.4, 75.9, 78.0, 78.0, 78.0, 78.0, 78.5, 79.2, 79.2, 79.2, 79.7, 79.7, 79.7, 80.2, 80.2, 80.2, 80.3, 80.4, 80.4, 81.1, 81.1, 81.1, 96.2, 98.8, 107.7 ] }, { "x": 16.0, "output_throughput": { "median": 160.77, "min": 157.69, "max": 160.84, "n": 3 }, "median_ttft_ms": { "median": 339.57, "min": 328.95, "max": 368.08, "n": 3 }, "p99_ttft_ms": { "median": 6402.25, "min": 6273.44, "max": 6529.05, "n": 3 }, "median_tpot_ms": { "median": 75.69, "min": 73.98, "max": 80.44, "n": 3 }, "p99_tpot_ms": { "median": 122.69, "min": 121.85, "max": 123.41, "n": 3 }, "goodput_rps": { "median": 0.63, "min": 0.62, "max": 0.63, "n": 3 }, "request_throughput": { "median": 1.26, "min": 1.23, "max": 1.26, "n": 3 }, "completed": [ 64, 64, 64 ], "ttft_quantiles_ms": [ 103.1, 108.1, 165.2, 167.4, 168.7, 170.7, 172.5, 172.9, 173.5, 174.2, 174.8, 175.3, 222.6, 224.8, 225.5, 226.5, 227.6, 228.7, 229.7, 232.2, 232.8, 233.2, 233.7, 234.0, 234.6, 235.5, 501.9, 2868.9, 2870.9, 3034.2, 3171.9, 3175.3, 3311.4, 3419.0, 3421.5, 3660.0, 3662.6, 5849.5, 5851.3, 6040.6, 6042.4, 6082.4, 6084.5, 6131.7, 6133.9, 6170.3, 6246.3, 6248.0, 6262.2, 6393.7, 6552.5 ], "tpot_quantiles_ms": [ 70.4, 70.4, 70.4, 70.4, 70.4, 70.4, 70.5, 70.8, 70.9, 70.9, 70.9, 70.9, 71.4, 71.4, 71.5, 71.9, 71.9, 73.6, 75.2, 75.2, 75.7, 75.7, 76.0, 76.0, 76.7, 78.2, 78.2, 78.3, 78.3, 79.3, 79.3, 80.0, 80.0, 80.4, 80.4, 80.6, 80.6, 80.9, 80.9, 81.4, 98.9, 99.2, 99.3, 100.5, 101.3, 101.3, 103.5, 103.5, 103.9, 121.4, 124.9 ] }, { "x": 32.0, "output_throughput": { "median": 193.24, "min": 189.21, "max": 194.53, "n": 3 }, "median_ttft_ms": { "median": 5785.29, "min": 5605.44, "max": 6216.5, "n": 3 }, "p99_ttft_ms": { "median": 11451.14, "min": 11082.67, "max": 11732.09, "n": 3 }, "median_tpot_ms": { "median": 117.94, "min": 113.38, "max": 118.67, "n": 3 }, "p99_tpot_ms": { "median": 174.91, "min": 172.06, "max": 176.33, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.51, "min": 1.48, "max": 1.52, "n": 3 }, "completed": [ 128, 128, 128 ], "ttft_quantiles_ms": [ 116.0, 371.6, 991.0, 1308.8, 1943.3, 2387.9, 2401.2, 2511.8, 2570.0, 2657.7, 2719.0, 2909.2, 3006.7, 3050.5, 3818.8, 4659.8, 4798.8, 5050.1, 5065.9, 5081.3, 5357.2, 5482.5, 5503.9, 5607.5, 5701.8, 5780.6, 6213.6, 6410.9, 6551.9, 6682.3, 7702.8, 7734.3, 7754.3, 7957.4, 7990.8, 8054.3, 8114.7, 8554.4, 8623.2, 8632.7, 10327.1, 10366.9, 10410.3, 10435.1, 10449.0, 10626.6, 10719.6, 10906.8, 10910.8, 11729.2, 12023.1 ], "tpot_quantiles_ms": [ 82.8, 83.3, 84.0, 86.5, 86.7, 88.1, 88.5, 90.7, 90.8, 93.0, 93.8, 95.6, 96.2, 97.1, 100.7, 100.8, 101.5, 103.3, 104.0, 107.2, 107.8, 110.8, 111.9, 113.4, 114.1, 117.5, 117.9, 118.0, 118.7, 119.7, 119.7, 124.5, 128.0, 128.8, 132.4, 132.6, 132.9, 133.0, 136.6, 138.4, 139.7, 141.3, 143.4, 148.8, 154.1, 155.3, 157.9, 162.0, 164.4, 165.8, 183.2 ] }, { "x": 64.0, "output_throughput": { "median": 200.98, "min": 193.35, "max": 201.84, "n": 3 }, "median_ttft_ms": { "median": 26858.23, "min": 26682.95, "max": 28855.16, "n": 3 }, "p99_ttft_ms": { "median": 36753.71, "min": 35295.1, "max": 36984.18, "n": 3 }, "median_tpot_ms": { "median": 120.85, "min": 120.56, "max": 128.03, "n": 3 }, "p99_tpot_ms": { "median": 173.38, "min": 168.53, "max": 174.26, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.57, "min": 1.51, "max": 1.58, "n": 3 }, "completed": [ 256, 256, 256 ], "ttft_quantiles_ms": [ 186.4, 2286.4, 2450.6, 4166.6, 5939.1, 6108.0, 8054.0, 8333.5, 10683.0, 10998.5, 11159.7, 11471.3, 11649.1, 11734.7, 12403.1, 12511.0, 12916.6, 13121.8, 13370.0, 21680.3, 24176.6, 25648.6, 25994.8, 26475.0, 26686.5, 26899.9, 27339.1, 28145.3, 28322.7, 28653.9, 29044.3, 29237.3, 29324.9, 29458.4, 29862.2, 29987.1, 30845.6, 31324.5, 31521.1, 31848.6, 31961.4, 32105.0, 32511.7, 33391.1, 33860.7, 34071.2, 34365.2, 34544.9, 34970.0, 36076.5, 37038.8 ], "tpot_quantiles_ms": [ 85.0, 86.3, 86.4, 89.5, 91.3, 93.4, 95.0, 95.3, 96.9, 97.2, 99.7, 99.8, 101.8, 104.7, 105.6, 106.4, 106.5, 108.1, 112.0, 116.1, 116.3, 118.4, 120.3, 120.8, 121.2, 125.8, 126.6, 128.0, 129.7, 129.9, 131.6, 137.0, 138.2, 139.8, 141.6, 141.9, 145.7, 146.9, 147.7, 150.3, 150.7, 156.0, 160.2, 161.4, 162.9, 163.0, 165.5, 168.0, 171.6, 173.4, 174.3 ] } ] } }, "rate": { "default": { "kv_pool_tokens": 24094, "points": [ { "x": 0.5, "output_throughput": { "median": 67.13, "min": 62.64, "max": 86.28, "n": 3 }, "median_ttft_ms": { "median": 602.31, "min": 480.02, "max": 602.32, "n": 3 }, "p99_ttft_ms": { "median": 1619.76, "min": 1403.92, "max": 1670.54, "n": 3 }, "median_tpot_ms": { "median": 96.81, "min": 95.84, "max": 107.03, "n": 3 }, "p99_tpot_ms": { "median": 134.46, "min": 108.08, "max": 153.86, "n": 3 }, "goodput_rps": { "median": 0.29, "min": 0.23, "max": 0.34, "n": 3 }, "request_throughput": { "median": 0.52, "min": 0.49, "max": 0.67, "n": 3 }, "completed": [ 32, 32, 32 ], "ttft_quantiles_ms": [ 70.9, 71.8, 243.4, 255.7, 296.4, 308.2, 322.3, 340.1, 371.0, 402.1, 421.3, 428.9, 438.4, 440.6, 447.0, 465.3, 473.2, 481.9, 485.8, 526.2, 530.3, 531.4, 544.1, 556.0, 589.0, 601.8, 602.7, 603.0, 614.0, 624.9, 633.2, 633.6, 645.5, 648.7, 654.2, 662.4, 668.4, 679.3, 699.1, 749.8, 847.2, 885.4, 986.3, 1007.0, 1067.3, 1130.1, 1194.9, 1329.6, 1533.0, 1613.5, 1696.2 ], "tpot_quantiles_ms": [ 73.3, 73.9, 77.1, 79.7, 80.7, 83.0, 83.8, 85.0, 86.2, 86.6, 87.5, 87.7, 87.9, 89.1, 89.1, 89.3, 89.8, 89.9, 90.5, 92.6, 94.1, 94.1, 94.5, 94.8, 95.5, 97.6, 98.2, 98.3, 99.3, 99.6, 100.2, 100.5, 101.9, 102.1, 104.1, 106.7, 107.4, 109.1, 111.1, 118.2, 119.2, 122.0, 123.2, 127.0, 128.3, 132.1, 133.0, 134.5, 136.4, 143.9, 155.0 ] }, { "x": 1.0, "output_throughput": { "median": 113.38, "min": 105.82, "max": 138.61, "n": 3 }, "median_ttft_ms": { "median": 203.78, "min": 185.76, "max": 243.58, "n": 3 }, "p99_ttft_ms": { "median": 310.2, "min": 285.52, "max": 373.4, "n": 3 }, "median_tpot_ms": { "median": 99.75, "min": 94.08, "max": 114.3, "n": 3 }, "p99_tpot_ms": { "median": 110.86, "min": 100.97, "max": 122.4, "n": 3 }, "goodput_rps": { "median": 0.44, "min": 0.18, "max": 0.91, "n": 3 }, "request_throughput": { "median": 0.89, "min": 0.83, "max": 1.08, "n": 3 }, "completed": [ 32, 32, 32 ], "ttft_quantiles_ms": [ 70.7, 71.3, 134.8, 142.9, 145.6, 151.1, 153.6, 157.9, 162.3, 163.8, 170.5, 171.9, 173.7, 176.4, 179.0, 179.3, 185.0, 185.3, 186.9, 190.5, 192.1, 196.4, 198.2, 199.9, 205.6, 207.8, 214.4, 219.0, 221.8, 226.1, 229.6, 234.1, 241.0, 241.8, 245.6, 247.9, 254.0, 256.3, 260.8, 268.6, 270.0, 270.2, 274.0, 275.6, 276.9, 282.3, 283.6, 298.3, 331.9, 344.8, 376.1 ], "tpot_quantiles_ms": [ 74.8, 75.2, 75.5, 75.8, 76.0, 76.4, 78.8, 86.5, 87.4, 89.0, 91.2, 92.2, 92.9, 93.9, 94.0, 94.4, 94.5, 95.5, 96.8, 97.7, 98.0, 98.0, 98.4, 98.7, 98.9, 99.3, 99.7, 100.2, 100.5, 100.9, 102.0, 102.8, 104.2, 105.4, 106.5, 108.7, 109.7, 110.0, 110.2, 110.9, 111.0, 111.6, 114.3, 115.2, 117.0, 117.5, 118.5, 119.6, 120.1, 121.1, 122.6 ] }, { "x": 2.0, "output_throughput": { "median": 167.34, "min": 164.76, "max": 176.02, "n": 3 }, "median_ttft_ms": { "median": 260.29, "min": 256.69, "max": 260.65, "n": 3 }, "p99_ttft_ms": { "median": 10372.53, "min": 7902.43, "max": 10548.43, "n": 3 }, "median_tpot_ms": { "median": 161.04, "min": 154.28, "max": 161.19, "n": 3 }, "p99_tpot_ms": { "median": 206.93, "min": 195.68, "max": 206.95, "n": 3 }, "goodput_rps": { "median": 0.04, "min": 0.0, "max": 0.11, "n": 3 }, "request_throughput": { "median": 1.31, "min": 1.29, "max": 1.38, "n": 3 }, "completed": [ 60, 60, 60 ], "ttft_quantiles_ms": [ 70.8, 111.5, 135.1, 138.8, 147.8, 150.0, 158.0, 164.0, 165.8, 167.9, 169.2, 173.1, 175.6, 179.7, 187.1, 188.6, 193.4, 195.2, 200.3, 202.5, 203.9, 213.6, 217.3, 223.5, 251.4, 260.0, 263.7, 477.4, 645.8, 799.2, 1076.3, 1291.3, 1385.2, 1573.6, 1763.6, 1947.5, 2536.5, 3471.8, 4073.0, 4289.0, 4708.3, 5009.3, 5356.6, 5524.8, 6091.2, 6943.0, 7378.5, 8431.2, 9440.9, 10220.4, 10936.5 ], "tpot_quantiles_ms": [ 88.1, 88.9, 92.6, 94.5, 96.9, 101.3, 104.5, 109.2, 109.6, 112.8, 116.0, 120.4, 122.5, 126.3, 129.6, 132.8, 134.8, 136.4, 136.6, 142.8, 143.3, 145.5, 146.7, 147.7, 154.9, 157.9, 160.0, 162.9, 164.6, 165.8, 167.3, 168.4, 172.0, 176.7, 178.9, 179.9, 182.5, 183.9, 185.9, 189.4, 190.4, 192.7, 194.4, 195.7, 196.8, 202.4, 203.8, 204.2, 205.5, 206.6, 207.5 ] }, { "x": 4.0, "output_throughput": { "median": 167.98, "min": 164.18, "max": 174.37, "n": 3 }, "median_ttft_ms": { "median": 20344.15, "min": 18973.91, "max": 23391.17, "n": 3 }, "p99_ttft_ms": { "median": 54485.89, "min": 48741.57, "max": 56505.72, "n": 3 }, "median_tpot_ms": { "median": 138.08, "min": 136.14, "max": 152.92, "n": 3 }, "p99_tpot_ms": { "median": 202.63, "min": 198.67, "max": 210.59, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.31, "min": 1.28, "max": 1.36, "n": 3 }, "completed": [ 120, 120, 120 ], "ttft_quantiles_ms": [ 70.5, 449.4, 703.0, 1666.5, 1901.0, 1969.3, 2347.3, 2860.8, 3154.7, 3332.1, 3699.2, 3967.5, 4166.2, 4469.5, 5442.8, 6103.4, 15284.9, 16816.9, 17209.8, 18304.3, 18928.3, 19191.4, 19334.8, 19853.0, 20083.9, 20393.3, 20883.9, 21194.7, 22273.0, 23465.3, 24208.5, 30852.0, 33377.1, 34000.2, 34378.1, 34736.6, 35435.9, 36799.6, 37775.9, 38423.4, 39005.2, 39535.1, 40587.5, 41161.8, 41969.0, 42177.7, 42681.3, 47378.4, 52344.4, 54130.4, 56570.8 ], "tpot_quantiles_ms": [ 80.5, 83.4, 87.4, 102.3, 104.3, 106.1, 108.8, 110.2, 113.0, 113.6, 115.9, 116.2, 116.9, 121.1, 121.1, 122.4, 124.3, 125.5, 131.4, 134.5, 136.1, 136.1, 138.1, 139.0, 140.0, 142.4, 146.1, 148.9, 152.5, 152.9, 155.4, 156.0, 157.8, 159.2, 160.6, 164.6, 165.5, 172.9, 173.7, 175.1, 175.8, 176.3, 179.0, 180.4, 180.5, 183.6, 191.1, 191.1, 200.2, 202.6, 212.7 ] }, { "x": 8.0, "output_throughput": { "median": 178.32, "min": 172.32, "max": 184.31, "n": 2 }, "median_ttft_ms": { "median": 64606.73, "min": 61124.18, "max": 68089.28, "n": 2 }, "p99_ttft_ms": { "median": 132295.5, "min": 125727.82, "max": 138863.18, "n": 2 }, "median_tpot_ms": { "median": 137.22, "min": 132.17, "max": 142.27, "n": 2 }, "p99_tpot_ms": { "median": 202.56, "min": 198.43, "max": 206.69, "n": 2 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 2 }, "request_throughput": { "median": 1.39, "min": 1.35, "max": 1.44, "n": 2 }, "completed": [ 240, 240 ], "ttft_quantiles_ms": [ 71.7, 1498.7, 2008.0, 3289.7, 3842.1, 5020.7, 5455.7, 7400.4, 19378.9, 21515.0, 21836.3, 23293.3, 23775.1, 25180.2, 25464.5, 26894.4, 41410.4, 42723.1, 44046.1, 44556.0, 45807.2, 46425.5, 48912.1, 50112.6, 61040.1, 62758.1, 65394.1, 66328.6, 69085.4, 71342.3, 72908.5, 78487.9, 83649.2, 86277.2, 88236.0, 89915.5, 93153.9, 95200.7, 97254.6, 104474.5, 106282.3, 107719.9, 108770.5, 114053.9, 117107.1, 118562.1, 119706.2, 123908.6, 125761.0, 137327.9, 139223.7 ], "tpot_quantiles_ms": [ 86.5, 90.8, 99.1, 102.0, 102.8, 103.7, 106.2, 106.7, 107.7, 110.0, 110.0, 110.7, 113.1, 114.0, 118.2, 118.9, 120.3, 121.6, 123.4, 124.0, 126.9, 130.8, 132.2, 132.7, 134.3, 134.3, 138.5, 138.9, 142.3, 143.0, 147.1, 150.0, 150.6, 151.7, 154.6, 158.3, 159.7, 160.4, 162.9, 165.0, 167.3, 170.5, 171.2, 171.7, 179.8, 180.3, 183.5, 188.2, 191.3, 199.6, 225.2 ] }, { "x": 16.0, "output_throughput": { "median": 181.01, "min": 175.23, "max": 186.78, "n": 2 }, "median_ttft_ms": { "median": 147567.56, "min": 141108.92, "max": 154026.2, "n": 2 }, "p99_ttft_ms": { "median": 295981.53, "min": 283713.66, "max": 308249.39, "n": 2 }, "median_tpot_ms": { "median": 135.54, "min": 133.49, "max": 137.6, "n": 2 }, "p99_tpot_ms": { "median": 205.07, "min": 198.65, "max": 211.49, "n": 2 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 2 }, "request_throughput": { "median": 1.41, "min": 1.37, "max": 1.46, "n": 2 }, "completed": [ 480, 480 ], "ttft_quantiles_ms": [ 71.0, 3135.3, 5240.6, 6983.3, 21898.9, 24265.3, 27087.2, 29636.3, 45126.8, 48129.0, 51163.6, 55560.8, 67463.3, 72984.4, 76055.1, 80597.7, 92825.7, 97394.8, 102066.1, 106467.6, 116320.1, 120487.7, 127643.1, 132128.9, 139249.3, 143782.2, 153988.8, 157153.5, 163984.0, 175217.5, 179153.7, 182797.1, 187619.1, 199910.2, 204254.7, 207507.5, 212604.2, 225553.5, 229990.0, 232269.6, 236515.2, 250975.1, 255668.3, 258085.7, 260676.5, 277290.0, 279384.5, 282727.2, 284048.5, 303696.0, 308614.9 ], "tpot_quantiles_ms": [ 87.8, 94.3, 97.7, 101.3, 102.2, 103.2, 106.1, 107.1, 108.1, 108.9, 109.5, 112.1, 112.2, 114.7, 115.0, 116.3, 118.4, 123.6, 125.3, 128.3, 129.3, 130.2, 130.6, 132.8, 134.7, 136.0, 137.5, 139.4, 145.4, 147.9, 148.9, 149.7, 151.8, 153.2, 154.7, 156.7, 157.9, 160.1, 161.2, 167.7, 170.2, 172.2, 173.0, 175.8, 178.2, 179.6, 181.3, 182.4, 184.4, 195.5, 220.1 ] } ] }, "fp8": { "kv_pool_tokens": 46228, "points": [ { "x": 0.5, "output_throughput": { "median": 60.17, "min": 59.77, "max": 79.08, "n": 3 }, "median_ttft_ms": { "median": 835.18, "min": 766.57, "max": 1374.19, "n": 3 }, "p99_ttft_ms": { "median": 1838.85, "min": 1559.22, "max": 3527.86, "n": 3 }, "median_tpot_ms": { "median": 173.36, "min": 162.11, "max": 188.45, "n": 3 }, "p99_tpot_ms": { "median": 219.53, "min": 203.67, "max": 231.09, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.03, "n": 3 }, "request_throughput": { "median": 0.47, "min": 0.47, "max": 0.62, "n": 3 }, "completed": [ 32, 32, 32 ], "ttft_quantiles_ms": [ 141.0, 157.2, 346.1, 405.5, 492.8, 534.5, 563.0, 592.8, 670.5, 672.8, 696.3, 708.2, 721.3, 722.1, 743.3, 752.0, 761.9, 773.4, 779.5, 787.9, 801.8, 821.1, 840.0, 851.2, 873.6, 904.4, 931.4, 980.4, 1002.6, 1027.7, 1057.0, 1079.0, 1149.4, 1170.2, 1175.2, 1247.5, 1280.5, 1321.8, 1426.6, 1484.0, 1507.2, 1508.1, 1560.3, 1695.7, 1830.7, 1866.9, 2381.8, 2692.4, 2832.1, 3150.5, 3552.9 ], "tpot_quantiles_ms": [ 96.0, 99.4, 118.0, 123.0, 130.5, 134.9, 140.3, 141.0, 145.2, 145.7, 150.0, 152.2, 152.2, 154.4, 154.4, 157.6, 158.4, 159.6, 161.2, 163.8, 168.3, 169.1, 169.4, 171.9, 173.6, 174.1, 175.3, 176.8, 178.4, 181.0, 186.2, 186.7, 186.9, 187.3, 188.7, 190.2, 192.9, 194.7, 195.5, 195.9, 198.8, 201.2, 202.7, 204.1, 206.4, 206.4, 212.6, 217.4, 219.1, 222.7, 232.5 ] }, { "x": 1.0, "output_throughput": { "median": 98.67, "min": 89.12, "max": 104.7, "n": 3 }, "median_ttft_ms": { "median": 289.49, "min": 261.9, "max": 312.47, "n": 3 }, "p99_ttft_ms": { "median": 441.79, "min": 418.58, "max": 500.66, "n": 3 }, "median_tpot_ms": { "median": 144.79, "min": 139.59, "max": 160.92, "n": 3 }, "p99_tpot_ms": { "median": 166.55, "min": 154.18, "max": 170.94, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.02, "n": 3 }, "request_throughput": { "median": 0.77, "min": 0.7, "max": 0.82, "n": 3 }, "completed": [ 32, 32, 32 ], "ttft_quantiles_ms": [ 59.7, 60.0, 67.9, 121.7, 135.5, 142.5, 144.4, 148.7, 189.5, 205.2, 209.5, 212.4, 216.9, 225.4, 227.4, 234.1, 245.8, 246.5, 251.8, 267.4, 271.7, 272.0, 281.4, 282.7, 287.4, 288.3, 290.3, 297.2, 302.4, 311.9, 315.0, 323.7, 326.3, 327.4, 330.0, 352.0, 356.2, 356.6, 362.9, 369.8, 377.6, 381.2, 390.8, 392.1, 403.7, 412.0, 421.8, 428.8, 439.8, 446.1, 510.9 ], "tpot_quantiles_ms": [ 92.7, 103.4, 106.8, 112.0, 118.0, 120.7, 121.4, 121.9, 125.4, 127.3, 129.6, 129.6, 130.1, 132.7, 133.9, 140.2, 141.8, 143.7, 144.5, 145.6, 145.8, 145.9, 146.4, 146.7, 148.2, 148.3, 148.4, 148.5, 148.7, 148.8, 150.7, 152.4, 153.0, 153.4, 155.2, 159.1, 162.5, 163.6, 164.1, 164.6, 164.8, 165.3, 165.4, 165.6, 166.1, 166.7, 167.0, 168.0, 168.8, 169.6, 170.9 ] }, { "x": 2.0, "output_throughput": { "median": 148.71, "min": 133.73, "max": 177.31, "n": 3 }, "median_ttft_ms": { "median": 457.49, "min": 369.84, "max": 477.23, "n": 3 }, "p99_ttft_ms": { "median": 11880.21, "min": 4935.89, "max": 12463.73, "n": 3 }, "median_tpot_ms": { "median": 278.95, "min": 256.78, "max": 333.87, "n": 3 }, "p99_tpot_ms": { "median": 290.42, "min": 264.39, "max": 346.59, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.16, "min": 1.04, "max": 1.39, "n": 3 }, "completed": [ 60, 60, 60 ], "ttft_quantiles_ms": [ 59.6, 84.5, 121.6, 135.3, 154.4, 159.1, 177.9, 189.3, 194.7, 207.9, 217.6, 233.4, 242.7, 265.1, 277.2, 278.2, 287.7, 301.3, 310.3, 330.3, 343.4, 361.6, 367.2, 378.5, 390.4, 426.4, 564.0, 604.9, 1076.0, 1885.5, 2432.5, 2603.0, 2817.2, 3005.3, 3185.3, 3414.1, 3457.0, 4110.4, 4613.6, 4787.5, 4977.5, 6875.3, 7785.7, 8227.7, 8717.4, 9081.7, 9323.1, 10024.9, 10914.5, 11868.8, 12659.1 ], "tpot_quantiles_ms": [ 128.1, 133.3, 138.0, 138.0, 156.7, 164.2, 172.2, 172.3, 172.3, 188.9, 188.9, 195.8, 195.8, 206.3, 206.3, 206.7, 230.5, 243.6, 243.9, 254.3, 256.8, 259.1, 259.3, 260.6, 261.5, 261.9, 262.3, 263.2, 264.2, 264.4, 275.6, 279.3, 283.5, 283.8, 283.8, 285.2, 285.9, 286.4, 288.1, 289.5, 309.3, 329.5, 334.1, 334.9, 335.1, 337.0, 341.7, 342.4, 342.8, 344.0, 347.0 ] }, { "x": 4.0, "output_throughput": { "median": 171.43, "min": 138.16, "max": 176.19, "n": 3 }, "median_ttft_ms": { "median": 652.92, "min": 573.07, "max": 691.81, "n": 3 }, "p99_ttft_ms": { "median": 46028.53, "min": 43567.71, "max": 60335.7, "n": 3 }, "median_tpot_ms": { "median": 345.01, "min": 340.32, "max": 420.1, "n": 3 }, "p99_tpot_ms": { "median": 488.54, "min": 486.77, "max": 617.58, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.34, "min": 1.08, "max": 1.38, "n": 3 }, "completed": [ 120, 120, 120 ], "ttft_quantiles_ms": [ 60.1, 104.7, 131.4, 157.5, 178.4, 202.5, 216.5, 231.8, 245.4, 249.5, 271.5, 275.9, 291.7, 305.1, 310.2, 317.8, 335.4, 345.4, 356.3, 363.2, 378.1, 391.2, 401.9, 420.3, 469.0, 551.3, 2208.2, 3613.1, 4772.1, 6066.5, 12272.7, 16674.5, 19328.6, 22260.0, 26153.9, 28535.3, 30658.3, 31331.5, 32143.0, 33911.9, 35126.8, 35416.6, 37092.1, 38341.4, 41038.8, 42626.2, 43616.3, 46434.4, 49691.3, 53679.3, 60687.6 ], "tpot_quantiles_ms": [ 131.7, 141.9, 161.9, 175.2, 184.2, 202.5, 213.4, 218.5, 220.5, 235.4, 245.3, 256.6, 261.0, 267.5, 280.4, 288.6, 290.5, 307.5, 315.5, 326.3, 328.7, 330.7, 341.2, 349.9, 355.8, 370.4, 371.4, 377.6, 386.7, 396.2, 401.4, 420.1, 420.1, 426.3, 433.3, 441.7, 447.9, 453.8, 457.4, 468.9, 480.3, 482.0, 484.6, 486.6, 488.5, 530.2, 553.4, 577.2, 601.2, 614.8, 618.7 ] }, { "x": 8.0, "output_throughput": { "median": 139.23, "min": 119.75, "max": 142.32, "n": 3 }, "median_ttft_ms": { "median": 77301.5, "min": 71103.87, "max": 93461.07, "n": 3 }, "p99_ttft_ms": { "median": 179415.32, "min": 174476.29, "max": 211238.2, "n": 3 }, "median_tpot_ms": { "median": 319.95, "min": 311.61, "max": 389.42, "n": 3 }, "p99_tpot_ms": { "median": 496.57, "min": 476.17, "max": 579.02, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.09, "min": 0.94, "max": 1.11, "n": 3 }, "completed": [ 240, 240, 240 ], "ttft_quantiles_ms": [ 60.1, 2895.3, 4814.1, 6841.6, 8532.9, 10379.6, 12015.6, 14585.1, 16010.4, 17308.3, 19925.8, 20798.5, 23238.9, 23851.1, 28740.3, 49366.8, 57353.2, 61602.6, 65662.2, 66855.1, 69178.8, 71155.5, 72963.2, 74746.0, 77497.9, 79248.3, 82760.4, 85017.3, 92701.5, 100270.6, 104549.5, 112793.5, 116321.9, 120926.6, 124043.6, 127134.8, 130265.4, 134111.4, 137115.6, 138434.6, 141511.8, 143345.9, 151238.7, 159913.1, 164057.1, 168733.4, 171681.6, 174204.2, 178797.8, 204568.6, 211438.5 ], "tpot_quantiles_ms": [ 114.6, 119.5, 149.2, 151.8, 188.2, 195.8, 204.3, 214.4, 220.8, 230.7, 232.0, 234.6, 246.2, 251.4, 267.5, 268.4, 270.1, 272.1, 288.3, 299.9, 302.3, 303.7, 310.3, 319.5, 327.9, 328.6, 339.8, 345.0, 353.9, 358.6, 359.0, 374.0, 379.8, 389.4, 390.5, 395.5, 406.6, 411.7, 422.6, 433.1, 435.9, 442.4, 456.0, 465.2, 476.2, 484.0, 496.3, 513.8, 545.0, 562.3, 600.2 ] }, { "x": 16.0, "output_throughput": { "median": 146.05, "min": 142.95, "max": 146.34, "n": 3 }, "median_ttft_ms": { "median": 188277.73, "min": 183974.65, "max": 193950.38, "n": 3 }, "p99_ttft_ms": { "median": 375048.38, "min": 374459.05, "max": 381324.92, "n": 3 }, "median_tpot_ms": { "median": 317.77, "min": 314.38, "max": 321.41, "n": 3 }, "p99_tpot_ms": { "median": 483.7, "min": 474.2, "max": 501.37, "n": 3 }, "goodput_rps": { "median": 0.0, "min": 0.0, "max": 0.0, "n": 3 }, "request_throughput": { "median": 1.14, "min": 1.12, "max": 1.14, "n": 3 }, "completed": [ 480, 480, 480 ], "ttft_quantiles_ms": [ 71.3, 4171.7, 8289.0, 11972.8, 15861.2, 19675.1, 23491.4, 26954.4, 56331.0, 64038.5, 68300.2, 72259.2, 76494.5, 81470.9, 87321.0, 112485.3, 120705.0, 125372.2, 130339.9, 136413.5, 140748.4, 146219.5, 149970.5, 178680.3, 184028.6, 188471.3, 193615.8, 197905.4, 202282.3, 205782.8, 211054.1, 241662.2, 246350.2, 250984.7, 254613.2, 260094.6, 263548.8, 269200.9, 295830.0, 302888.2, 307871.1, 312468.3, 316763.2, 320540.7, 324747.9, 336083.3, 358786.5, 365674.7, 370353.8, 374609.9, 382546.9 ], "tpot_quantiles_ms": [ 126.7, 154.1, 186.8, 193.1, 196.2, 203.2, 209.9, 216.3, 220.1, 223.5, 230.2, 233.8, 236.2, 247.6, 253.5, 256.7, 263.9, 267.5, 269.9, 282.5, 287.6, 291.6, 297.2, 300.8, 307.1, 314.8, 321.4, 325.4, 333.7, 338.4, 346.7, 350.1, 354.4, 364.7, 372.4, 376.5, 381.5, 388.1, 397.2, 401.6, 411.9, 416.7, 425.1, 431.9, 435.9, 448.0, 458.7, 463.5, 468.6, 483.7, 534.3 ] } ] } }, "quantiles": [ 0.0, 0.02, 0.04, 0.06, 0.08, 0.1, 0.12, 0.14, 0.16, 0.18, 0.2, 0.22, 0.24, 0.26, 0.28, 0.3, 0.32, 0.34, 0.36, 0.38, 0.4, 0.42, 0.44, 0.46, 0.48, 0.5, 0.52, 0.54, 0.56, 0.58, 0.6, 0.62, 0.64, 0.66, 0.68, 0.7, 0.72, 0.74, 0.76, 0.78, 0.8, 0.82, 0.84, 0.86, 0.88, 0.9, 0.92, 0.94, 0.96, 0.98, 1.0 ] }