Hpt
Home
GitHub
crate.io
Benchmarks
Home
GitHub
crate.io
Benchmarks
  • Benchmarks

    • unary
    • binary
    • reduce
    • conv(f32)
    • conv(f16)
    • pooling
    • normalization
    • matmul
    • fft
    • nn

      • resnet(f32)
      • resnet(f16)
      • lstm

Matmul Benchmark

::: chartjs I5 12600k Matmul f32 Performance (M = N = K)

{
  "type": "line",
  "data": {
    "labels": [64, 128, 192, 256, 320, 384, 448, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1088, 1152, 1216, 1280, 1344, 1408, 1472, 1536, 1600, 1664, 1728, 1792, 1856, 1920, 1984, 2048, 2112, 2176, 2240, 2304, 2368, 2432, 2496, 2560, 2624, 2688, 2752, 2816, 2880, 2944, 3008, 3072, 3136, 3200, 3264, 3328, 3392, 3456, 3520, 3584, 3648, 3712, 3776, 3840, 3904, 3968, 4032, 4096, 4160, 4224, 4288, 4352, 4416, 4480, 4544, 4608, 4672, 4736, 4800, 4864, 4928, 4992, 5056, 5120],
    "datasets": [
      {
        "label": "faer-gemm",
        "data": [109.01046258254462, 141.44295075687063, 202.87419207686062, 288.0919217245592, 346.993171484019, 394.9180694544829, 464.3924131062779, 550.1657329567887, 506.0350519010833, 573.1686163026682, 626.1361478746818, 647.2475914028324, 668.9760529655205, 671.498489788445, 677.3978676203661, 669.9406527266975, 665.6851606299348, 666.4936059788569, 678.7648542195485, 684.3691541229308, 680.3479368990405, 683.0635246713998, 673.993411186458, 674.2507103905853, 668.3166637138742, 666.1467526037414, 669.9318560062217, 665.4447931812235, 670.4264078747034, 669.7767922732487, 673.7887016875015, 664.2146653044796, 638.5207582427936, 652.1348885942981, 657.7508249147671, 656.8019428812275, 653.5011743985076, 659.4600630256335, 661.122652041117, 659.2562741109308, 663.7608903198862, 664.1527732976928, 658.618538038762, 663.7252127495012, 644.4870861936852, 659.9716915306667, 672.7261143859141, 674.7398295732379, 671.0244188091782, 664.0006315868313, 674.2985490177902, 678.1884738178472, 681.4996325227445, 684.5479737313433, 687.198655480231, 679.6754040777639, 685.05033574518, 683.4847232988088, 677.6447146129641, 682.4116179572161, 689.2754968317405, 683.9217868856049, 686.369850973822, 688.917060010025, 678.043757946786, 673.5062683109918, 682.1822701449275, 679.1057648444902, 679.9578625819186, 677.8393667546173, 666.9573213719566, 678.5348385020804, 674.0150194844679, 682.9125056637736, 681.8249075215783, 675.3223271361503, 682.8952168445078, 688.0584042477876, 690.2388016875835, 666.7547342275211],
        "backgroundColor": "rgba(255, 206, 86)",
        "borderColor": "rgb(255, 206, 86)",
        "borderWidth": 3,
        "pointRadius": 0
      },
      {
        "label": "Hpt",
        "data": [58.82442700128803, 141.47381827834727, 180.84143955947368, 273.73268716478293, 325.39410640252225, 380.65632389643787, 438.1154651283244, 526.9823930443115, 617.3605578746237, 629.6386700425104, 664.0163523942367, 671.719449246938, 678.7435934134606, 704.5144397406939, 709.9009028453527, 702.0171553312083, 701.8353669321244, 704.0635962526735, 690.5212894429211, 679.0609920593863, 688.8405340711486, 683.3737820198386, 678.7746067698248, 681.219001089079, 676.1394329456282, 683.5968759643916, 684.7396908518974, 685.7588050116708, 684.3755023219707, 675.1144305233083, 689.972412454375, 682.115809710711, 667.3334936087979, 667.6383900352195, 667.3624676743394, 675.6020414195586, 671.3297849138019, 652.7173956146567, 673.73353314143, 664.7348111283427, 673.9869435228469, 679.1121201017614, 681.4963656703821, 680.2415382802715, 685.0525676003168, 671.704521597304, 678.2343737173999, 672.6659591480544, 674.742269639805, 677.6794155319385, 679.5695815575228, 674.2599430449469, 681.8965454897526, 679.1046254159829, 679.3490342679128, 682.8679461825183, 688.2878136699367, 679.9260901030243, 681.5047161518987, 685.0950272232304, 686.8883955440115, 683.3607353787258, 687.2694182752293, 670.2704387807852, 689.4067129518793, 682.6571686956521, 688.2864764033174, 688.3211875407097, 686.3252703407053, 692.5891931446178, 694.6083374717749, 692.5834274429304, 696.336445530898, 689.1147600129743, 693.3667711598746, 699.5436142492401, 688.1965885681426, 697.3147953363228, 696.0000840926224, 696.5113025428127],
        "backgroundColor": "rgb(67, 241, 32)",
        "borderColor": "rgb(67, 241, 32)",
        "borderWidth": 3,
        "pointRadius": 0
      },
      {
        "label": "Candle(mkl)",
        "data": [16.385940709852825, 80.02149014627953, 157.61139469586286, 263.51423283665173, 348.77660761423573, 500.4378502767966, 535.2046774714664, 411.2365061886212, 430.02532744478583, 484.008440089194, 508.32698708277195, 547.2449906983359, 573.7719783914262, 578.8183362367064, 607.0293901283156, 594.2188153749438, 607.7882595262228, 601.4183113552058, 602.8718428327701, 612.2362187919643, 618.197524108963, 612.2139246806335, 614.1590437589356, 611.7122423540835, 612.6019735416774, 626.8352960884683, 625.4279214545454, 617.565228835597, 632.660355460819, 627.4916057529172, 633.0719644722453, 625.9344785707942, 632.2321065826571, 624.4183218159229, 628.8132296394965, 637.8944462006051, 645.517559486366, 643.8015279177939, 620.4416068481133, 636.3301391663841, 637.5401906406657, 636.8320912412443, 653.3365053254084, 648.602993632838, 663.6238207801739, 654.3883978414755, 657.384607889124, 619.6039891410603, 655.1772383152194, 661.1645092114913, 675.7791399726171, 660.9601354411524, 657.208177719416, 626.3769774810319, 634.2347261423711, 618.63400723446, 664.2722981650107, 652.1828514886835, 653.7810877474195, 634.6102998038666, 633.1652808087257, 628.37571266784, 630.4238592738639, 620.3518549853305, 636.1059951402696, 634.1215938073202, 628.1076747420833, 633.3189566500191, 649.4469328506788, 646.6407191657677, 648.9657353207679, 644.9882907844429, 658.7756618087853, 644.5815549514563, 649.5859030837005, 634.5460410476977, 666.1697008182576, 649.6133654725849, 645.4292914656678, 606.4967374604608],
        "backgroundColor": "rgb(234, 32, 241)",
        "borderColor": "rgb(234, 32, 241)",
        "borderWidth": 3,
        "pointRadius": 0
      }
    ]
  },
  "options": {
    "animation": false,
    "responsive": true,
    "plugins": {
      "legend": {
        "position": "top"
      }
    },
    "scales": {
      "y": {
        "beginAtZero": true,
        "title": {
          "display": true,
          "text": "GFlops"
        }
      },
      "x": {
        "beginAtZero": true,
        "title": {
          "display": true,
          "text": "N"
        }
      }
    }
  }
}

:::

::: chartjs Ryzen R9 9950x3d Matmul f32 Performance (M = N = K)

{
  "type": "line",
  "data": {
    "labels": [64, 128, 192, 256, 320, 384, 448, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1088, 1152, 1216, 1280, 1344, 1408, 1472, 1536, 1600, 1664, 1728, 1792, 1856, 1920, 1984, 2048, 2112, 2176, 2240, 2304, 2368, 2432, 2496, 2560, 2624, 2688, 2752, 2816, 2880, 2944, 3008, 3072, 3136, 3200, 3264, 3328, 3392, 3456, 3520, 3584, 3648, 3712, 3776, 3840, 3904, 3968, 4032, 4096, 4160, 4224, 4288, 4352, 4416, 4480, 4544, 4608, 4672, 4736, 4800, 4864, 4928, 4992, 5056, 5120],
    "datasets": [
      {
        "label": "Hpt(faer-gemm)",
        "data": [157.91046214640258, 182.4259236545867, 323.6313863398234, 423.90996255177527, 531.2118147940254, 646.9911987968588, 736.7481431532724, 1170.5723661141733, 950.2900428011727, 1136.916652235323, 1320.8379561482316, 1415.0371000476912, 1486.8772779193573, 1541.4551987660013, 1719.1732374306748, 1759.7045869845604, 1632.9930890018395, 1705.998984830359, 1772.300872481644, 1809.054241638116, 1818.2566475139668, 1913.18131814045, 1936.8451573933287, 1921.5773802449805, 1917.9941063215917, 1930.1252591191212, 1930.3595223189363, 1943.8607231161036, 2006.8153828616973, 2004.2194584186275, 1988.0451486556374, 2006.924465943024, 2011.0660032312228, 2029.8188535352087, 2046.8181452727251, 2060.8242498513887, 2035.507191338964, 2066.4107255621507, 2101.0910028156736, 2086.6509966753883, 2103.460895758548, 2105.8053969639723, 2080.805331292709, 2104.671593518688, 2113.8090145270107, 2129.487039926057, 2164.4332199543655, 2163.4397533530428, 2122.781279948596, 2117.257699307424, 2174.7661697715703, 2172.4404713682143, 2173.2745223695847, 2183.133935339561, 2184.4693895978803, 2193.443731157015, 2172.830156722555, 2179.0759344922963, 2196.3989450037625, 2214.4800304296523, 2206.260785033467, 2231.087864043803, 2230.5200192158964, 2239.271988311847, 2244.0813667272946, 2255.39528167489, 2262.4012901657898, 2253.288454612182, 2272.0517640471717, 2267.0219988238587, 2272.4875063760924, 2286.3691811694903, 2279.807908341661, 2280.6943174205576, 2297.969074715262, 2293.3971400182922, 2298.904522796091, 2291.3225405969442, 2305.9271296342554, 2314.0987586206898],
        "backgroundColor": "rgba(255, 206, 86)",
        "borderColor": "rgb(255, 206, 86)",
        "borderWidth": 3,
        "pointRadius": 0
      },
      {
        "label": "Hpt",
        "data": [128.28751939160523, 98.08780069367417, 171.56983673094902, 280.2348216305024, 423.2077970560531, 575.7524892151466, 720.6136974830964, 697.3394066586545, 918.0437095853584, 1035.6244597933355, 1190.7595422509153, 1290.7424368295056, 1339.0070573738572, 1464.3018060817442, 1495.9589953137656, 1540.8069508881497, 1568.1576184163653, 1699.6425119798168, 1710.5829890650573, 1675.3480272275528, 1777.8696090077017, 1748.43164910802, 1751.3094784368118, 1737.4760685402316, 1833.1005494788265, 1827.1861488699503, 1844.661005703264, 1897.9285497144867, 1872.7110189722584, 1875.8245531236773, 1893.454049387459, 1944.55050111431, 1961.666312207073, 1964.0733591835578, 2049.9209997199373, 2011.0088960272071, 2008.3980173743296, 2025.2260895252023, 2070.9287583781065, 2091.888320007965, 2096.487133635353, 2161.556406381163, 2146.868533830279, 2140.6563511507834, 2182.0152757324895, 2202.371623908901, 2185.652416295536, 2166.238963398249, 2261.779530678982, 2194.7947111848052, 2236.5286568263623, 2302.880322652998, 2327.366284050912, 2323.664653802828, 2326.0105389363493, 2355.2868106761894, 2340.7244846780304, 2349.5020968006083, 2409.878324964797, 2402.9204485479563, 2453.4815570825867, 2480.8785789821427, 2537.375612079572, 2540.644215821243, 2455.952235886459, 2463.8611044570534, 2519.266172445073, 2587.0613419569418, 2602.528778097227, 2581.9448756052407, 2612.920912410504, 2573.9814572685455, 2585.558899837427, 2606.507791217862, 2608.381780788177, 2599.457033371681, 2638.9343585440856, 2667.6898323677447, 2661.6125812037035, 2661.472806771236],
        "backgroundColor": "rgb(67, 241, 32)",
        "borderColor": "rgb(67, 241, 32)",
        "borderWidth": 3,
        "pointRadius": 0
      }
    ]
  },
  "options": {
    "animation": false,
    "responsive": true,
    "plugins": {
      "legend": {
        "position": "top"
      }
    },
    "scales": {
      "y": {
        "beginAtZero": true,
        "title": {
          "display": true,
          "text": "GFlops"
        }
      },
      "x": {
        "beginAtZero": true,
        "title": {
          "display": true,
          "text": "N"
        }
      }
    }
  }
}

:::

::: chartjs I5 12600k Matmul f16 Performance (M = N = K)

{
  "type": "line",
  "data": {
    "labels": [64, 128, 192, 256, 320, 384, 448, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1088, 1152, 1216, 1280, 1344, 1408, 1472, 1536, 1600, 1664, 1728, 1792, 1856, 1920, 1984, 2048, 2112, 2176, 2240, 2304, 2368, 2432, 2496, 2560, 2624, 2688, 2752, 2816, 2880, 2944, 3008, 3072, 3136, 3200, 3264, 3328, 3392, 3456, 3520, 3584, 3648, 3712, 3776, 3840, 3904, 3968, 4032, 4096, 4160, 4224, 4288, 4352, 4416, 4480, 4544, 4608, 4672, 4736, 4800, 4864, 4928, 4992, 5056, 5120],
    "datasets": [
      {
        "label": "faer-gemm",
        "data": [63.29320527121452, 62.55929203645419, 147.59738599600198, 223.16800442259893, 269.18826937852816, 324.78542508077885, 340.2204012426476, 291.202171777519, 321.2399028377893, 351.89858430570223, 421.6469590988618, 435.71549245713715, 481.255393495723, 466.0971787710845, 502.23229538501903, 528.3429274865817, 485.97260704949224, 504.41577759078916, 521.2462465915315, 566.5244186075206, 586.394219247582, 571.5738417520812, 605.0255364964852, 605.8945076659389, 593.0093073194445, 618.8532670955275, 608.9402820566823, 623.14531064028, 629.0494648172115, 630.2247728123726, 645.7855269506446, 586.2397194553917, 620.0457367165525, 635.9811605946524, 635.485294317822, 645.1484607763826, 650.217979926563, 658.9696267631564, 661.5427139506618, 661.8400696587469, 671.9360751355528, 657.6397536802777, 655.9802949115893, 660.1069318640159, 655.9559730733973, 661.5940079611735, 662.6644716195708, 663.2920077204697, 656.6265050193285, 660.8874220997759, 661.6517021673933, 663.1402138092012, 660.7315850299303, 653.4814166293795, 667.7347486774111, 672.0683314452554, 666.5524090282253, 673.2140852648897, 680.8583316598165, 685.7172752043596, 684.1242571313596, 688.0644849339207, 688.7136408510638, 691.3428243058349, 673.4452385406921, 675.9224342959641, 673.0108055655143, 681.2104314710743, 679.4214066745561, 687.6894225621414, 691.4091465291084, 684.4681616789087, 683.7309584177003, 687.9989653886009, 687.1202236719479, 679.9109278818315, 677.0997836039604, 683.7095877328936, 681.6836266666668, 679.2395141700405],
        "backgroundColor": "rgba(255, 206, 86)",
        "borderColor": "rgb(255, 206, 86)",
        "borderWidth": 3,
        "pointRadius": 0
      },{
        "label": "Hpt",
        "data": [12.61442990999592, 57.886921664411304, 131.8317168884913, 196.18234561699413, 254.3876592174226, 298.6421572005272, 341.3074655173824, 319.8417378562101, 389.8732182028379, 442.3664716238406, 479.3831403581477, 513.865029964978, 525.5298986723834, 548.705308229679, 562.6005203318445, 567.6103289975927, 575.3822121433425, 597.6433130930226, 606.6434600175515, 615.223179127741, 608.5377389461598, 615.1132370814163, 605.1050404296766, 604.1366944650204, 612.5543817640903, 611.967927268934, 618.2502814877961, 610.6712399389003, 617.0387858390612, 621.6936491799083, 628.3093722981052, 622.1955351342295, 663.5791131551783, 663.1182703382135, 667.3881997661207, 668.2671391365905, 672.8539781078092, 667.4477379871311, 668.2437967846201, 672.9414137941161, 669.4785611894944, 677.2054520234423, 677.6690278101717, 683.4048236125141, 668.884739043719, 678.0562277847271, 683.2761864803307, 681.8293378340644, 687.9958913672681, 690.5890080921475, 680.5714125200903, 686.3974590689013, 682.6917601598452, 693.55770625297, 690.4623957965822, 694.0203121031653, 693.5319113142856, 683.0283569133534, 687.597350906769, 699.0506666666666, 683.3385847143268, 662.1754661579228, 645.1606374803149, 648.755975794194, 697.0834761558945, 694.7716194883614, 698.964679716312, 707.5232807553648, 707.6143245357437, 715.6020055710305, 713.3565571868465, 702.5289801615507, 707.4469125771765, 710.5487642541807, 713.7270087124879, 711.4369369026274, 702.5382257235103, 695.5602990662566, 699.0114419469983, 703.8160880964867],
        "backgroundColor": "rgb(67, 241, 32)",
        "borderColor": "rgb(67, 241, 32)",
        "borderWidth": 3,
        "pointRadius": 0
      },
      {
        "label": "Candle(mkl)",
        "data": [16.96542636842902, 101.53359322499993, 227.2577838547775, 322.9152032744954, 380.9822337734332, 464.6624117619616, 516.7756341135262, 547.5622287857096, 573.4994617261693, 524.801964804231, 534.2993446529254, 411.64886991308833, 402.9274088604431, 371.46482281239537, 374.6280611735998, 390.0580392618429, 399.1074706634945, 411.53446325447413, 410.7597823639808, 413.5052811720643, 438.8948629232683, 445.37862551689096, 442.2689820880514, 441.95392041330615, 437.17414403071683, 457.4569415716954, 469.68091633743046, 456.54056231853906, 467.9601818595369, 474.44709317800766, 465.15873646937706, 474.3437776468703, 489.99696534230605, 499.0122960306427, 501.3966270082747, 503.1551347130756, 512.0112486430762, 514.5387665007177, 515.1261552021358, 527.219373988636, 523.1166616356468, 544.4418230845794, 543.9082869905253, 547.7165406929817, 542.1380995335961, 545.5477663208782, 548.3910483889875, 555.8538974432596, 550.0769225371604, 565.7784172824684, 569.1270661865793, 582.4526239298801, 588.2021445064055, 583.4380610035336, 591.2454767417813, 597.8789701818182, 600.6462578657593, 589.4259882224143, 578.7570284977156, 613.9669720791542, 611.5283377595067, 616.5926990574883, 601.6367211381367, 576.0224370159262, 606.3701495051589, 608.3983969646821, 626.6101003139281, 608.2011599926212, 623.8077746903296, 628.120097799511, 638.9119590330268, 628.2165246356341, 632.0326770870778, 630.053619549229, 646.3588544710695, 625.7472786514409, 630.5447141833508, 638.6086216016425, 638.2578548938271, 638.3720713436386],
        "backgroundColor": "rgb(234, 32, 241)",
        "borderColor": "rgb(234, 32, 241)",
        "borderWidth": 3,
        "pointRadius": 0
      }
    ]
  },
  "options": {
    "animation": false,
    "responsive": true,
    "plugins": {
      "legend": {
        "position": "top"
      }
    },
    "scales": {
      "y": {
        "beginAtZero": true,
        "title": {
          "display": true,
          "text": "GFlops"
        }
      },
      "x": {
        "beginAtZero": true,
        "title": {
          "display": true,
          "text": "N"
        }
      }
    }
  }
}

:::

最近更新: 2025/6/24 21:23
Contributors: Jianqoq
Prev
normalization
Next
fft