Model eXplorer#

At MemryX, we do not maintain a curated model zoo which limits customers choice. Instead, we recommend customers choose/train their own model(s) and use MemryX automated tools to generate executable DFP files (.dfp = “dataflow program”).

Our automated tools do not modify original source models (e.g. no pruning, no distillation, etc.) and do not require any retraining (or any dataset) to maintain high accuracy. Instead, our tools compile models as they were trained, resulting in dataflow programs that execute with high accuracy and high utilization of available hardware.

The list of models below was created using a model explorer that automatically aggregates, compiles, and tests models from various public online sources on MemryX hardware. Anyone can use these publicly available models, access links for detailed model source information, and either recreate or simply download the compiled DFP. Unless explicitly stated, none of these models were developed by or are owned by MemryX. This model list will evolve over time depending on customer feedback.

Of course, a customer can certainly choose to modify any given model(s) prior to using MemryX tools. For example, if a customer is willing to tradeoff accuracy for performance, one can choose to prune, quantize, and/or distill model(s) and then create an executable for measuring performance & accuracy.

Benchmark Details

  • SDK Version: 2.1

  • Compilation: Closed-loop method is used to create DFPs. See Closed Loop Compilation tutorial.

  • Host: Intel i7-9700T (1.5 GHz), 16 GB RAM

  • Metrics: FPS and latency reported for Default (14 TFLOPS) and Full (20 TFLOPS) modes. Benchmarks up to 40 TFLOPS will be added in future releases.

  • Latency 1: Frame-to-frame latency (used to compute FPS)

  • Latency 2: Full system latency (including host) from initial input to final output

Understanding Model Icons
Tutorial code

Models featured in step-by-step coding tutorials. See Tutorials.

Example app (video)

Used in a complete example application that processes video streams. Browse MemryX eXamples.

Example app (image)

Used in a complete example application for still-image workflows. Browse MemryX eXamples.

Example app (audio)

Used in a complete example application for audio/voice pipelines. Browse MemryX eXamples.

Needs Neural Compiler extension

Requires additional extensions to run properly. See Neural Compiler and Neural Compiler Extensions tutorial.

MXA optimized

Optimized/retrained by the MemryX team for better MXA performance while preserving accuracy.

Models (Below is a partial list of tested models)

Below is a partial list of AI models that MemryX has downloaded, compiled, and benchmarked from public sources. This is not a model zoo; the models remain unmodified (unless explicitly mentioned), and MemryX customers are not restricted to using these models. Customers are welcome to compile additional models as needed. For guidance, see the Neural Compiler and Tutorials pages.

Classification (ImageNet)
Classification (ImageNet)
Object Detection
Object Detection
Segmentation
Segmentation
Pose Estimation
Pose Estimation
Image Enhancement/Super Resolution
Image Enhancement/Super Resolution
Face/Head
Face/Head
Classification (Other)
Classification (Other)
ADAS/Driver Monitoring
ADAS/Driver Monitoring
Hand/Palm
Hand/Palm
Depth Estimation
Depth Estimation
Point Cloud
Point Cloud
Audio/Speech
Audio/Speech
Emotion Detection
Emotion Detection
Style Transfer/GANs
Style Transfer/GANs
Gesture Recognition
Gesture Recognition
Tracking
Tracking
Wireframe/keypoint Detection
Wireframe/keypoint Detection
Re-identification
Re-identification
Classification (ImageNet)

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

1

ConvNext-tiny

224×224×3

27.3 M

ONNX

376

529

2.7 ms

1.9 ms

18.2 ms

13.1 ms

HuggingFace

DFP

2

CS3-DarkNet

256×256×3

8.9 M

ONNX

1681

2019

0.6 ms

0.5 ms

2.5 ms

1.9 ms

Timm

DFP

3

CS3-SE-DarkNet

256×256×3

20.9 M

ONNX

110

156

9.1 ms

6.4 ms

31.5 ms

22.4 ms

Timm

DFP

4

CSP-DarkNet-50

256×256×3

26.3 M

ONNX

467

662

2.1 ms

1.5 ms

9.8 ms

7.1 ms

OpenMMLab

DFP

5

CSP-DarkNet-53

256×256×3

26.3 M

ONNX

468

663

2.1 ms

1.5 ms

9.8 ms

7.1 ms

Timm

DFP

6

CSP-ResNet-50

256×256×3

20.6 M

ONNX

868

1126

1.2 ms

0.9 ms

5.3 ms

4.0 ms

OpenMMLab

DFP

7

DarkNet-53

256×256×3

39.7 M

ONNX

82

117

12.2 ms

8.5 ms

27.4 ms

19.5 ms

Timm

DFP

8

DarkNet-AA53

256×256×3

34.3 M

ONNX

224

317

4.5 ms

3.2 ms

14.2 ms

10.2 ms

Timm

DFP

9

DenseNet121

224×224×3

7.8 M

ONNX

737

1036

1.4 ms

1.0 ms

10.6 ms

7.6 ms

ONNX Zoo

DFP

10

DenseNet121

224×224×3

7.7 M

Keras

711

1007

1.4 ms

1.0 ms

11.0 ms

7.9 ms

Keras.Applications

DFP

11

DenseNet169

224×224×3

13.6 M

Keras

521

737

1.9 ms

1.4 ms

18.9 ms

13.5 ms

Keras.Applications

DFP

12

DenseNet201

224×224×3

19.3 M

Keras

333

472

3.0 ms

2.1 ms

32.0 ms

22.8 ms

Keras.Applications

DFP

13

DLA34

224×224×3

15.0 M

ONNX

603

854

1.7 ms

1.2 ms

4.4 ms

3.3 ms

OpenVINO

DFP

14

EfficientNet-B0

224×224×3

5.0 M

ONNX

1013

1433

1.0 ms

0.7 ms

7.4 ms

5.5 ms

OpenVINO

DFP

15

EfficientNet-B1

240×240×3

7.4 M

ONNX

518

735

1.9 ms

1.4 ms

14.6 ms

10.5 ms

OpenMMLab

DFP

16

EfficientNet-EdgeTpu(L)

300×300×3

10.1 M

TFLite

165

234

6.1 ms

4.3 ms

14.4 ms

10.3 ms

Coral

DFP

17

EfficientNet-EdgeTpu(M)

240×240×3

6.5 M

TFLite

449

636

2.2 ms

1.6 ms

6.7 ms

4.9 ms

Coral

DFP

18

EfficientNet-EdgeTpu(S)

224×224×3

3.9 M

TFLite

487

690

2.1 ms

1.4 ms

4.2 ms

3.1 ms

Coral

DFP

19

EfficientNet-lite0

224×224×3

4.4 M

TFLite

1948

2317

0.5 ms

0.4 ms

2.0 ms

1.6 ms

Kaggle

DFP

20

EfficientNet-lite1

240×240×3

5.1 M

TFLite

1938

2309

0.5 ms

0.4 ms

2.5 ms

1.9 ms

Kaggle

DFP

21

EfficientNet-lite2

260×260×3

5.8 M

TFLite

1538

1781

0.7 ms

0.6 ms

3.4 ms

2.6 ms

Kaggle

DFP

22

EfficientNet-lite3

280×280×3

7.8 M

TFLite

1218

1724

0.8 ms

0.6 ms

4.3 ms

3.2 ms

Kaggle

DFP

23

EfficientNet-lite4

300×300×3

12.4 M

TFLite

750

1062

1.3 ms

0.9 ms

8.9 ms

6.4 ms

Kaggle

DFP

24

EfficientNet-lite4

224×224×3

12.4 M

ONNX

1403

1985

0.7 ms

0.5 ms

6.0 ms

4.4 ms

ONNX Zoo

DFP

25

EfficientNet-v2-B0

224×224×3

6.8 M

ONNX

1100

1558

0.9 ms

0.6 ms

6.6 ms

4.8 ms

OpenMMLab

DFP

26

EfficientNet-v2-B0

224×224×3

6.9 M

Keras

1102

1518

0.9 ms

0.7 ms

6.1 ms

4.5 ms

Keras.Applications

DFP

27

EfficientNet-v2-B1

240×240×3

7.7 M

ONNX

1016

1438

1.0 ms

0.7 ms

10.0 ms

7.2 ms

OpenMMLab

DFP

28

EfficientNet-v2-B1

240×240×3

7.8 M

Keras

1018

1442

1.0 ms

0.7 ms

8.6 ms

6.2 ms

Keras.Applications

DFP

29

EfficientNet-v2-B2

260×260×3

9.6 M

ONNX

761

1077

1.3 ms

0.9 ms

14.6 ms

10.5 ms

OpenMMLab

DFP

30

EfficientNet-v2-B2

260×260×3

9.7 M

Keras

762

1080

1.3 ms

0.9 ms

13.4 ms

9.6 ms

Keras.Applications

DFP

31

EfficientNet-v2-B3

300×300×3

13.8 M

Keras

524

742

1.9 ms

1.3 ms

25.3 ms

18.0 ms

Keras.Applications

DFP

32

FastViT

256×256×3

3.8 M

ONNX

1125

1591

0.9 ms

0.6 ms

3.8 ms

2.8 ms

Timm

DFP

33

FBNet-v3

224×224×3

8.2 M

ONNX

1783

2274

0.6 ms

0.4 ms

7.8 ms

5.7 ms

Timm

DFP

34

FBNetC

224×224×3

5.3 M

ONNX

1945

2311

0.5 ms

0.4 ms

2.1 ms

1.6 ms

Timm

DFP

35

GerNet-large

256×256×3

29.6 M

ONNX

488

691

2.0 ms

1.4 ms

9.3 ms

6.8 ms

Timm

DFP

36

GhostNet

224×224×3

4.9 M

ONNX

1625

2250

0.6 ms

0.4 ms

3.7 ms

2.8 ms

Timm

DFP

37

GhostNet-v2

224×224×3

5.8 M

ONNX

1033

1463

1.0 ms

0.7 ms

6.2 ms

4.6 ms

Timm

DFP

38

GoogleNet

224×224×3

6.7 M

ONNX

518

734

1.9 ms

1.4 ms

3.8 ms

2.8 ms

ONNX Zoo

DFP

39

HardCoRe-NAS

224×224×3

4.9 M

ONNX

1937

2310

0.5 ms

0.4 ms

2.5 ms

2.0 ms

Timm

DFP

40

HBONet0.25

224×224×3

1.8 M

ONNX

1954

2314

0.5 ms

0.4 ms

1.4 ms

1.1 ms

OpenVINO

DFP

41

HBONet1.0

224×224×3

4.3 M

ONNX

1113

1178

0.9 ms

0.8 ms

3.1 ms

2.5 ms

OpenVINO

DFP

42

HGNet-tiny

224×224×3

14.0 M

ONNX

372

527

2.7 ms

1.9 ms

9.7 ms

7.0 ms

Timm

DFP

43

HGNet-v2

224×224×3

37.7 M

ONNX

96

136

10.4 ms

7.4 ms

41.1 ms

29.2 ms

Timm

DFP

44

Inception

224×224×3

6.3 M

TFLite

753

1066

1.3 ms

0.9 ms

2.9 ms

2.3 ms

Coral

DFP

45

Inception

224×224×3

6.7 M

ONNX

537

760

1.9 ms

1.3 ms

3.7 ms

2.8 ms

ONNX Zoo

DFP

46

Inception-ResNet-v2

299×299×3

28.9 M

TensorFlow

168

238

6.0 ms

4.2 ms

36.4 ms

25.9 ms

OpenVINO

DFP

47

Inception-v2

224×224×3

10.7 M

ONNX

751

1063

1.3 ms

0.9 ms

3.5 ms

2.7 ms

ONNX Zoo

DFP

48

Inception-v3

299×299×3

22.7 M

TFLite

317

449

3.2 ms

2.2 ms

10.0 ms

7.2 ms

Kaggle

DFP

49

Inception-v3

346×346×3

22.7 M

TFLite

275

390

3.6 ms

2.6 ms

11.9 ms

8.6 ms

AI Benchmark

DFP

50

Inception-v3

299×299×3

22.7 M

TensorFlow

317

449

3.2 ms

2.2 ms

10.0 ms

7.2 ms

OpenVINO

DFP

51

Inception-v3

299×299×3

22.7 M

ONNX

317

449

3.2 ms

2.2 ms

10.0 ms

7.2 ms

OpenMMLab

DFP

52

Inception-v3

224×224×3

10.7 M

TFLite

750

1062

1.3 ms

0.9 ms

3.6 ms

2.7 ms

Coral

DFP

53

Inception-v3

299×299×3

22.7 M

Keras

317

449

3.2 ms

2.2 ms

10.0 ms

7.2 ms

Keras.Applications

DFP

54

LCNet-0.5

224×224×3

1.8 M

ONNX

1947

2309

0.5 ms

0.4 ms

1.2 ms

1.0 ms

Timm

DFP

55

MobileNet

128×128×3

4.0 M

TFLite

6822

7920

0.1 ms

0.1 ms

1.0 ms

0.8 ms

Kaggle

DFP

56

MobileNet

224×224×3

4.0 M

TFLite

1956

2321

0.5 ms

0.4 ms

1.4 ms

1.2 ms

Kaggle

DFP

57

MobileNet

192×192×3

4.0 M

TFLite

3585

4221

0.3 ms

0.2 ms

1.3 ms

1.0 ms

Kaggle

DFP

58

MobileNet

224×224×3

4.0 M

ONNX

1954

2324

0.5 ms

0.4 ms

1.4 ms

1.2 ms

OpenCV

DFP

59

MobileNet

160×160×3

4.0 M

TFLite

4266

4895

0.2 ms

0.2 ms

1.2 ms

1.0 ms

Kaggle

DFP

60

MobileNet

224×224×3

4.1 M

Keras

1952

2321

0.5 ms

0.4 ms

1.4 ms

1.2 ms

Keras.Applications

DFP

61

MobileNet-EdgeTPU-v2-extra-small

224×224×3

4.2 M

TFLite

1945

2316

0.5 ms

0.4 ms

1.8 ms

1.4 ms

Kaggle

DFP

62

MobileNet-EdgeTPU-v2-m4

224×224×3

8.0 M

ONNX

564

799

1.8 ms

1.3 ms

4.2 ms

3.1 ms

Timm

DFP

63

MobileNet-v1

224×224×3

4.1 M

TensorFlow

1941

2321

0.5 ms

0.4 ms

1.6 ms

1.3 ms

OpenVINO

DFP

64

MobileNet-v2

224×224×3

3.3 M

TFLite

1950

2328

0.5 ms

0.4 ms

1.5 ms

1.2 ms

Coral

DFP

65

MobileNet-v2

224×224×3

3.3 M

ONNX

1952

2321

0.5 ms

0.4 ms

1.5 ms

1.2 ms

OpenVINO

DFP

66

MobileNet-v2

224×224×3

3.4 M

Keras

1952

2319

0.5 ms

0.4 ms

1.5 ms

1.2 ms

Keras.Applications

DFP

67

MobileNet-v2-0.5

224×224×3

1.9 M

ONNX

1955

2323

0.5 ms

0.4 ms

1.1 ms

0.9 ms

Timm

DFP

68

MobileNet-v3

512×512×3

5.2 M

TFLite

453

579

2.2 ms

1.7 ms

9.6 ms

7.2 ms

AI Benchmark

DFP

69

MobileNet-v3

224×224×3

3.9 M

TFLite

1937

2318

0.5 ms

0.4 ms

1.9 ms

1.5 ms

Coral

DFP

70

MobileNet-v4

224×224×3

4.0 M

ONNX

1952

2315

0.5 ms

0.4 ms

1.4 ms

1.2 ms

Timm

DFP

71

MobileNet-v4-conv-small

224×224×3

3.6 M

ONNX

1956

2321

0.5 ms

0.4 ms

1.3 ms

1.1 ms

Timm

DFP

72

MobileOne

224×224×3

7.4 M

ONNX

1935

2315

0.5 ms

0.4 ms

2.3 ms

1.8 ms

GitHub/PINTO

DFP

73

Mobileone-s4

224×224×3

14.3 M

ONNX

818

1158

1.2 ms

0.9 ms

11.1 ms

8.0 ms

Timm

DFP

74

NasNet(Mobile)

224×224×3

4.2 M

TFLite

1947

2320

0.5 ms

0.4 ms

1.9 ms

1.5 ms

Kaggle

DFP

75

NasNet(Mobile)

160×160×3

4.2 M

TFLite

4259

4880

0.2 ms

0.2 ms

1.5 ms

1.2 ms

Kaggle

DFP

76

NasNet(Mobile)

192×192×3

4.2 M

TFLite

3574

4213

0.3 ms

0.2 ms

1.7 ms

1.3 ms

Kaggle

DFP

77

NasNet(Mobile)

224×224×3

4.2 M

ONNX

1951

2318

0.5 ms

0.4 ms

1.8 ms

1.4 ms

Timm

DFP

78

NasNet(Mobile)

128×128×3

4.2 M

TFLite

4734

6694

0.2 ms

0.1 ms

1.4 ms

1.1 ms

Kaggle

DFP

79

NasNet(Mobile)-1.3

224×224×3

5.9 M

TFLite

1947

2312

0.5 ms

0.4 ms

2.4 ms

1.8 ms

Kaggle

DFP

80

NasNet(Mobile)-small

224×224×3

1.9 M

ONNX

1939

2317

0.5 ms

0.4 ms

2.0 ms

1.6 ms

Timm

DFP

81

PP-LCNet-v2

224×224×3

6.3 M

TensorFlow

1948

2315

0.5 ms

0.4 ms

2.2 ms

1.7 ms

GitHub/PINTO

DFP

82

PpResNet-50

224×224×3

24.4 M

ONNX

741

1050

1.3 ms

1.0 ms

6.0 ms

4.4 ms

OpenCV

DFP

83

RegNetX002

224×224×3

2.6 M

ONNX

1947

2308

0.5 ms

0.4 ms

2.1 ms

1.6 ms

Timm

DFP

84

RegNetX002

224×224×3

2.6 M

Keras

1948

2320

0.5 ms

0.4 ms

1.9 ms

1.5 ms

Keras.Applications

DFP

85

RegNetX004

224×224×3

5.0 M

Keras

1871

2272

0.5 ms

0.4 ms

3.4 ms

2.5 ms

Keras.Applications

DFP

86

RegNetX006

224×224×3

5.9 M

Keras

1198

1697

0.8 ms

0.6 ms

3.8 ms

2.9 ms

Keras.Applications

DFP

87

RegNetX008

224×224×3

7.0 M

Keras

1003

1420

1.0 ms

0.7 ms

4.7 ms

3.4 ms

Keras.Applications

DFP

88

RegNetX016

224×224×3

8.8 M

Keras

289

410

3.5 ms

2.4 ms

12.2 ms

8.8 ms

Keras.Applications

DFP

89

RegNetY002

224×224×3

3.0 M

ONNX

1922

2297

0.5 ms

0.4 ms

5.0 ms

3.7 ms

Timm

DFP

90

RegNetY002

224×224×3

3.0 M

Keras

1928

2303

0.5 ms

0.4 ms

3.8 ms

2.9 ms

Keras.Applications

DFP

91

RegNetY004

224×224×3

4.2 M

Keras

773

1094

1.3 ms

0.9 ms

10.2 ms

7.4 ms

Keras.Applications

DFP

92

RegNetY006

224×224×3

5.8 M

Keras

1118

1583

0.9 ms

0.6 ms

9.4 ms

6.8 ms

Keras.Applications

DFP

93

RegNetY008

224×224×3

6.0 M

Keras

1286

1769

0.8 ms

0.6 ms

7.8 ms

5.7 ms

Keras.Applications

DFP

94

RegNetY016

224×224×3

10.7 M

Keras

220

311

4.5 ms

3.2 ms

57.3 ms

40.6 ms

Keras.Applications

DFP

95

RepGhostNet-0.5

224×224×3

2.2 M

ONNX

1954

2313

0.5 ms

0.4 ms

1.9 ms

1.6 ms

Timm

DFP

96

RepVGG-A0

224×224×3

7.9 M

ONNX

1840

2320

0.5 ms

0.4 ms

2.0 ms

1.6 ms

OpenMMLab

DFP

97

RepVGG-A1

224×224×3

12.2 M

ONNX

1941

2315

0.5 ms

0.4 ms

2.1 ms

1.6 ms

OpenMMLab

DFP

98

RepVGG-A2

224×224×3

24.3 M

ONNX

742

1051

1.3 ms

1.0 ms

4.8 ms

3.5 ms

OpenMMLab

DFP

99

RepVGG-B0

224×224×3

13.7 M

ONNX

1079

1527

0.9 ms

0.7 ms

3.2 ms

2.4 ms

OpenMMLab

DFP

100

RepViT

224×224×3

5.2 M

ONNX

1923

2290

0.5 ms

0.4 ms

5.0 ms

3.7 ms

Timm

DFP

101

Res2NeXt50

224×224×3

23.5 M

ONNX

333

473

3.0 ms

2.1 ms

14.1 ms

10.1 ms

Timm

DFP

102

ResNet32-TS

256×256×3

17.1 M

ONNX

852

1206

1.2 ms

0.8 ms

3.7 ms

2.8 ms

Timm

DFP

103

ResNet33-TS

256×256×3

18.7 M

ONNX

804

1139

1.2 ms

0.9 ms

3.8 ms

2.8 ms

Timm

DFP

104

ResNet-18

224×224×3

11.1 M

ONNX

1836

2238

0.5 ms

0.4 ms

1.8 ms

1.4 ms

OpenMMLab

DFP

105

ResNet-18-v2

224×224×3

11.2 M

ONNX

1832

2238

0.5 ms

0.4 ms

1.8 ms

1.4 ms

ONNX Zoo

DFP

106

ResNet-34

224×224×3

20.8 M

ONNX

898

1272

1.1 ms

0.8 ms

4.5 ms

3.3 ms

OpenMMLab

DFP

107

ResNet-34-v2

224×224×3

20.8 M

ONNX

967

1369

1.0 ms

0.7 ms

4.4 ms

3.3 ms

ONNX Zoo

DFP

108

ResNet-50

224×224×3

24.3 M

TFLite

811

1148

1.2 ms

0.9 ms

4.9 ms

3.6 ms

Coral

DFP

109

ResNet-50

224×224×3

24.4 M

ONNX

825

1169

1.2 ms

0.9 ms

6.2 ms

4.6 ms

ONNX Zoo

DFP

110

ResNet-50

224×224×3

24.4 M

Keras

811

1148

1.2 ms

0.9 ms

5.2 ms

3.8 ms

Keras.Applications

DFP

111

ResNet-50

224×224×3

24.4 M

Keras

811

1148

1.2 ms

0.9 ms

5.2 ms

3.8 ms

Keras.Applications

DFP

112

ResNet-50-v2

224×224×3

24.4 M

ONNX

691

979

1.4 ms

1.0 ms

6.5 ms

4.8 ms

ONNX Zoo

DFP

113

ResNet-50-v2

224×224×3

24.4 M

Keras

820

1162

1.2 ms

0.9 ms

5.7 ms

4.2 ms

Keras.Applications

DFP

114

ResNet-B-26

224×224×3

15.2 M

ONNX

1210

1476

0.8 ms

0.7 ms

2.3 ms

1.8 ms

Timm

DFP

115

ResNet-B-50 (Blur)

224×224×3

24.4 M

ONNX

531

752

1.9 ms

1.3 ms

7.7 ms

5.5 ms

Timm

DFP

116

ResNet-Q-51

256×256×3

34.0 M

ONNX

138

196

7.2 ms

5.1 ms

15.8 ms

11.4 ms

Timm

DFP

117

ResNet-v1-d50

224×224×3

24.4 M

ONNX

720

1019

1.4 ms

1.0 ms

5.5 ms

4.0 ms

OpenMMLab

DFP

118

ResNeXt

256×256×3

9.8 M

ONNX

904

976

1.1 ms

1.0 ms

2.9 ms

2.4 ms

Timm

DFP

119

ReXNet

224×224×3

4.6 M

ONNX

1929

2298

0.5 ms

0.4 ms

4.4 ms

3.3 ms

Timm

DFP

120

ReXNetr-200

224×224×3

15.7 M

ONNX

774

1097

1.3 ms

0.9 ms

10.9 ms

7.8 ms

Timm

DFP

121

SelecSLS60

224×224×3

29.2 M

ONNX

322

457

3.1 ms

2.2 ms

7.5 ms

5.5 ms

Timm

DFP

122

SeNasNet(Mobile)

224×224×3

3.7 M

ONNX

1944

2316

0.5 ms

0.4 ms

2.8 ms

2.2 ms

Timm

DFP

123

SeNasNet(Mobile)-0.75

224×224×3

2.8 M

ONNX

1946

2315

0.5 ms

0.4 ms

2.4 ms

1.9 ms

Timm

DFP

124

SeResNet33-TS

256×256×3

18.8 M

ONNX

304

391

3.3 ms

2.6 ms

17.2 ms

12.9 ms

Timm

DFP

125

ShuffleNet

224×224×3

1.4 M

ONNX

1950

2314

0.5 ms

0.4 ms

2.0 ms

1.5 ms

ONNX Zoo

DFP

126

ShuffleNet-v2

224×224×3

2.2 M

ONNX

1953

2321

0.5 ms

0.4 ms

1.7 ms

1.3 ms

ONNX Zoo

DFP

127

SPNasNet

224×224×3

4.2 M

ONNX

1945

2319

0.5 ms

0.4 ms

2.0 ms

1.6 ms

Timm

DFP

128

SqueezeNet

224×224×3

1.2 M

TFLite

1035

1466

1.0 ms

0.7 ms

1.8 ms

1.4 ms

Kaggle

DFP

129

SqueezeNet

224×224×3

1.2 M

ONNX

1960

2325

0.5 ms

0.4 ms

1.1 ms

0.9 ms

ONNX Zoo

DFP

130

SqueezeNet1.1

224×224×3

1.2 M

ONNX

1954

2325

0.5 ms

0.4 ms

1.1 ms

0.9 ms

ONNX Zoo

DFP

131

TinyNet-A

192×192×3

5.9 M

ONNX

1339

1895

0.7 ms

0.5 ms

7.2 ms

5.3 ms

Timm

DFP

132

VoVNet-v2

224×224×3

6.2 M

ONNX

1118

1234

0.9 ms

0.8 ms

3.8 ms

3.2 ms

Timm

DFP

133

Xception

299×299×3

25.7 M

ONNX

258

365

3.9 ms

2.7 ms

15.9 ms

11.4 ms

Timm

DFP

134

Xception

299×299×3

21.8 M

Keras

300

425

3.3 ms

2.4 ms

13.0 ms

9.3 ms

Keras.Applications

DFP

135

YOLO-v8-large-CLS

224×224×3

35.7 M

TFLite

258

366

3.9 ms

2.7 ms

13.4 ms

9.6 ms

Ultralytics

DFP

136

YOLO-v8-medium-CLS

224×224×3

16.3 M

TFLite

1018

1441

1.0 ms

0.7 ms

4.3 ms

3.2 ms

Ultralytics

DFP

137

YOLO-v8-nano-CLS

224×224×3

2.6 M

TFLite

1957

2322

0.5 ms

0.4 ms

1.3 ms

1.1 ms

Ultralytics

DFP

138

YOLO-v8-small-CLS

224×224×3

6.1 M

TFLite

1951

2318

0.5 ms

0.4 ms

1.7 ms

1.4 ms

Ultralytics

DFP

139

YOLO-v11-large-CLS

224×224×3

13.4 M

ONNX

477

676

2.1 ms

1.5 ms

9.3 ms

6.7 ms

Ultralytics

DFP

140

YOLO-v11-medium-CLS

224×224×3

11.1 M

ONNX

661

936

1.5 ms

1.1 ms

4.4 ms

3.3 ms

Ultralytics

DFP

141

YOLO-v11-nano-CLS

224×224×3

2.7 M

ONNX

1940

2319

0.5 ms

0.4 ms

1.7 ms

1.3 ms

Ultralytics

DFP

142

YOLO-v11-small-CLS

224×224×3

6.4 M

ONNX

1951

2307

0.5 ms

0.4 ms

2.5 ms

2.0 ms

Ultralytics

DFP

143

YOLO-v11-x-CLS

224×224×3

28.2 M

ONNX

229

324

4.4 ms

3.1 ms

19.3 ms

13.8 ms

Ultralytics

DFP
Object Detection

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

144

CenterNet(DLA)

512×512×3

17.1 M

ONNX

109

155

9.2 ms

6.5 ms

21.1 ms

15.4 ms

OpenVINO

DFP

145

CenterNet-MobileNet-v2

320×320×3

2.3 M

TensorFlow

839

941

1.2 ms

1.1 ms

3.8 ms

3.1 ms

TF Object Detection Zoo

DFP

146

CenterNet-MobileNet-v2-kpts

320×320×3

2.3 M

ONNX

777

805

1.3 ms

1.2 ms

3.8 ms

3.3 ms

TF Object Detection Zoo

DFP

147

CenterNet-ResNet-fpn

512×512×3

24.2 M

TensorFlow

66

94

15.2 ms

10.6 ms

26.6 ms

19.4 ms

TF Object Detection Zoo

DFP

148

DAMO-YOLO-medium

640×640×3

26.9 M

ONNX

35

50

28.6 ms

20.0 ms

48.8 ms

34.5 ms

DaMoYolo

DFP

149

DAMO-YOLO-Nl

416×416×3

5.4 M

ONNX

312

442

3.2 ms

2.3 ms

9.9 ms

7.2 ms

DaMoYolo

DFP

150

DAMO-YOLO-Ns

416×416×3

1.3 M

ONNX

611

865

1.6 ms

1.2 ms

4.6 ms

3.5 ms

DaMoYolo

DFP

151

DAMO-YOLO-small

640×640×3

15.5 M

ONNX

150

212

6.7 ms

4.7 ms

19.1 ms

13.7 ms

DaMoYolo

DFP

152

DAMO-YOLO-tiny

640×640×3

8.1 M

ONNX

386

386

2.6 ms

2.6 ms

10.0 ms

10.0 ms

DaMoYolo

DFP

153

DbNet-ResNet-18

736×1344×3

11.6 M

ONNX

16

23

62.5 ms

43.5 ms

74.0 ms

52.8 ms

OpenMMLab

DFP

154

DroNet

608×608×3

67.4 K

TFLite

364

446

2.7 ms

2.2 ms

3.4 ms

2.9 ms

GitHub/PINTO

DFP

155

EfficientDet-Lite1

384×384×3

4.2 M

TFLite

217

222

4.6 ms

4.5 ms

15.0 ms

12.3 ms

Kaggle

DFP

156

License-plate-Detection-LPD-YuNet

240×320×3

1008.7 K

ONNX

725

914

1.4 ms

1.1 ms

2.3 ms

2.0 ms

OpenCV

DFP

157

MediaPipe-object-Detection-3d-chair

224×224×3

2.4 M

TFLite

1945

2291

0.5 ms

0.4 ms

1.9 ms

1.5 ms

MediaPipe

DFP

158

MediaPipe-object-Detection-3d-chair-1stag…

640×480×3

1.4 M

TFLite

412

504

2.4 ms

2.0 ms

5.2 ms

4.0 ms

MediaPipe

DFP

159

MediaPipe-object-Detection-3d-cup

224×224×3

2.4 M

TFLite

1948

2298

0.5 ms

0.4 ms

1.9 ms

1.5 ms

MediaPipe

DFP

160

MediaPipe-object-Detection-3d-sneakers

224×224×3

2.4 M

TFLite

1939

2309

0.5 ms

0.4 ms

1.9 ms

1.5 ms

MediaPipe

DFP

161

Mobile-object-labeler

224×224×3

3.5 M

TFLite

1953

2323

0.5 ms

0.4 ms

1.4 ms

1.2 ms

Kaggle

DFP

162

Mobile-object-localizer

192×192×3

1.7 M

TFLite

3585

4205

0.3 ms

0.2 ms

1.3 ms

1.0 ms

GitHub/PINTO

DFP

163

MobileNet-YOLO-v4-syg

416×416×3

10.0 M

TensorFlow

766

959

1.3 ms

1.0 ms

7.9 ms

5.9 ms

OpenVINO

DFP

164

NanoDet

416×416×3

919.8 K

ONNX

786

947

1.3 ms

1.1 ms

3.5 ms

2.9 ms

OpenCV

DFP

165

NanoDet+ (ShuffleNet)-v2-1.5x

416×416×3

2.3 M

ONNX

616

863

1.6 ms

1.2 ms

5.9 ms

4.5 ms

OpenVINO

DFP

166

NanoDetwithShuffleNet-v2-1.5x

416×416×3

2.0 M

ONNX

641

892

1.6 ms

1.1 ms

4.4 ms

3.5 ms

OpenVINO

DFP

167

Person-Detection

512×512×3

1.7 M

TFLite

475

582

2.1 ms

1.7 ms

3.6 ms

2.9 ms

GitHub/PINTO

DFP

168

Person-vehicle-bike-Detection-crossroad-0…

1024×1024×3

1.1 M

TFLite

117

155

8.5 ms

6.5 ms

12.9 ms

9.9 ms

GitHub/PINTO

DFP

169

Person-vehicle-bike-Detection-crossroad-1…

512×512×3

2.8 M

TFLite

475

580

2.1 ms

1.7 ms

4.2 ms

3.4 ms

GitHub/PINTO

DFP

170

PP-YOLOE-Plus

640×640×3

7.6 M

ONNX

73

103

13.7 ms

9.7 ms

29.8 ms

21.6 ms

GitHub/PINTO

DFP

171

RtmDet-small

640×640×3

9.6 M

ONNX

56

79

17.9 ms

12.7 ms

39.1 ms

27.8 ms

OpenMMLab

DFP

172

RtmDet-tiny

640×640×3

5.3 M

ONNX

128

182

7.8 ms

5.5 ms

16.8 ms

12.4 ms

OpenMMLab

DFP

173

RtmDet-tiny-fast

640×640×3

5.3 M

ONNX

126

179

7.9 ms

5.6 ms

16.9 ms

12.5 ms

OpenMMLab

DFP

174

SpaghettiNet-EdgeTPU

320×320×3

3.3 M

TFLite

1188

1421

0.8 ms

0.7 ms

3.2 ms

2.6 ms

GitHub/PINTO

DFP

175

SSD300

300×300×3

32.7 M

ONNX

52

74

19.2 ms

13.5 ms

28.3 ms

20.2 ms

OpenMMLab

DFP

176

SSD512

512×512×3

34.4 M

ONNX

20

29

50.0 ms

34.5 ms

64.0 ms

45.4 ms

OpenMMLab

DFP

177

SSD-FPNMobileNet

640×640×3

29.5 M

TFLite

41

58

24.4 ms

17.2 ms

39.6 ms

28.8 ms

Coral

DFP

178

SSD-MobileNet

300×300×3

6.5 M

TFLite

1200

1443

0.8 ms

0.7 ms

2.0 ms

1.7 ms

Coral

DFP

179

SSD-MobileNet-v2

300×300×3

5.8 M

TFLite

1197

1442

0.8 ms

0.7 ms

2.2 ms

1.8 ms

Coral

DFP

180

SSD-MobileNet-v2 (MNAS-FPN)

320×320×3

2.4 M

TFLite

309

313

3.2 ms

3.2 ms

7.0 ms

6.2 ms

GitHub/PINTO

DFP

181

SSDlite-MobileNet-v2

320×320×3

4.0 M

ONNX

1198

1419

0.8 ms

0.7 ms

2.4 ms

1.9 ms

OpenMMLab

DFP

182

SSDlite-MobileNet-v2

300×300×3

4.3 M

TensorFlow

1197

1431

0.8 ms

0.7 ms

2.4 ms

1.9 ms

GitHub/PINTO

DFP

183

SSDLiteMobileDet

320×320×3

3.9 M

TFLite

1195

1419

0.8 ms

0.7 ms

3.2 ms

2.6 ms

Coral

DFP

184

SSSD-ResNet34

1200×1200×3

19.0 M

ONNX

21

30

47.6 ms

33.3 ms

62.1 ms

44.4 ms

OpenVINO

DFP

185

Text-Detection-db

480×640×3

11.6 M

TensorFlow

59

83

16.9 ms

12.0 ms

22.6 ms

16.8 ms

GitHub/PINTO

DFP

186

Vehicle-license-plate-detection

300×300×3

619.4 K

TFLite

1206

1452

0.8 ms

0.7 ms

1.5 ms

1.3 ms

GitHub/PINTO

DFP

187

Vehicle-license-plate-detection

256×256×3

561.2 K

TensorFlow

1689

2016

0.6 ms

0.5 ms

2.3 ms

1.9 ms

OpenVINO

DFP

188

YOLO-NAS-small

640×640×3

12.3 M

ONNX

106

150

9.4 ms

6.7 ms

20.5 ms

14.6 ms

Ultralytics

DFP

189

YOLO-tiny

416×416×3

15.1 M

TensorFlow

492

697

2.0 ms

1.4 ms

3.9 ms

3.0 ms

GitHub/RedisAI

DFP

190

YOLO-v2-tiny

416×416×3

15.1 M

TFLite

593

841

1.7 ms

1.2 ms

3.3 ms

2.5 ms

OpenVINO

DFP

191

YOLO-v3-lite

256×256×3

6.4 M

ONNX

1667

2009

0.6 ms

0.5 ms

2.7 ms

2.1 ms

GitHub/PINTO

DFP

192

YOLO-v3-MobileNet-v2

320×320×3

3.2 M

ONNX

713

1009

1.4 ms

1.0 ms

4.4 ms

3.5 ms

OpenMMLab

DFP

193

YOLO-v3-MobileNet-v2

416×416×3

3.2 M

ONNX

546

614

1.8 ms

1.6 ms

5.6 ms

4.7 ms

OpenMMLab

DFP

194

YOLO-v4-tiny

416×416×3

5.8 M

TFLite

608

675

1.6 ms

1.5 ms

3.3 ms

2.9 ms

AI Benchmark

DFP

195

YOLO-v5-medium-v6.1-fast

640×640×3

20.2 M

ONNX

106

150

9.4 ms

6.7 ms

19.5 ms

14.0 ms

OpenMMLab

DFP

196

YOLO-v5-medium-v6.1-fastb64-VOC

512×512×3

20.0 M

ONNX

186

256

5.4 ms

3.9 ms

15.2 ms

11.0 ms

OpenMMLab

DFP

197

YOLO-v5-mu

640×640×3

23.9 M

ONNX

96

137

10.4 ms

7.3 ms

26.2 ms

18.7 ms

Ultralytics

DFP

198

YOLO-v5-n6u

1280×1280×3

4.1 M

ONNX

83

97

12.0 ms

10.3 ms

23.1 ms

20.5 ms

Ultralytics

DFP

199

YOLO-v5-nano-p6-v6.2-fast

1280×1280×3

3.1 M

ONNX

66

66

15.2 ms

15.2 ms

28.8 ms

27.1 ms

OpenMMLab

DFP

200

YOLO-v5-nano-v6.1-fast

640×640×3

1.8 M

ONNX

271

272

3.7 ms

3.7 ms

7.5 ms

6.9 ms

OpenMMLab

DFP

201

YOLO-v5-nano-v6.1-fastb64-VOC

512×512×3

1.7 M

ONNX

475

573

2.1 ms

1.7 ms

3.8 ms

3.2 ms

OpenMMLab

DFP

202

YOLO-v5-nu

640×640×3

2.5 M

ONNX

326

389

3.1 ms

2.6 ms

6.3 ms

5.4 ms

Ultralytics

DFP

203

YOLO-v5-s6u

1280×1280×3

14.6 M

ONNX

42

59

23.8 ms

16.9 ms

43.1 ms

31.0 ms

Ultralytics

DFP

204

YOLO-v5-small-p6-v6.2-fast

1280×1280×3

12.0 M

ONNX

46

66

21.7 ms

15.2 ms

40.4 ms

32.3 ms

OpenMMLab

DFP

205

YOLO-v5-small-v6.1-fast

640×640×3

6.9 M

ONNX

269

269

3.7 ms

3.7 ms

8.5 ms

8.5 ms

OpenMMLab

DFP

206

YOLO-v5-small-v6.1-fastb64-VOC

512×512×3

6.7 M

ONNX

474

573

2.1 ms

1.7 ms

5.3 ms

4.1 ms

OpenMMLab

DFP

207

YOLO-v5-su

640×640×3

8.7 M

ONNX

266

376

3.8 ms

2.7 ms

9.5 ms

7.2 ms

Ultralytics

DFP

208

YOLO-v6-medium-fast

640×640×3

32.7 M

ONNX

56

80

17.9 ms

12.5 ms

45.7 ms

32.5 ms

OpenMMLab

DFP

209

YOLO-v6-nano-fast

640×640×3

4.1 M

ONNX

327

393

3.1 ms

2.5 ms

6.5 ms

5.1 ms

OpenMMLab

DFP

210

YOLO-v6-small-fast

640×640×3

16.4 M

ONNX

143

202

7.0 ms

5.0 ms

16.3 ms

11.7 ms

OpenMMLab

DFP

211

YOLO-v6-tiny-fast

640×640×3

9.2 M

ONNX

228

323

4.4 ms

3.1 ms

10.4 ms

7.8 ms

OpenMMLab

DFP

212

YOLO-v7

256×640×3

35.2 M

ONNX

57

81

17.5 ms

12.3 ms

54.1 ms

38.4 ms

GitHub/PINTO

DFP

213

YOLO-v7-13M (MXA-Optimized)

640×640×3

12.4 M

ONNX

85

121

11.8 ms

8.3 ms

27.3 ms

21.0 ms

GitHub/Official YOLOv7

DFP

214

YOLO-v7-22M (MXA-Optimized)

640×640×3

20.4 M

ONNX

59

84

16.9 ms

11.9 ms

40.1 ms

29.0 ms

GitHub/Official YOLOv7

DFP

215

YOLO-v7-27M (MXA-Optimized)

640×640×3

25.5 M

ONNX

48

68

20.8 ms

14.7 ms

50.4 ms

35.9 ms

GitHub/Official YOLOv7

DFP

216

YOLO-v7-large-fast

640×640×3

35.2 M

ONNX

22

31

45.5 ms

32.3 ms

101.5 ms

71.9 ms

OpenMMLab

DFP

217

YOLO-v7-tiny

320×320×3

5.9 M

ONNX

1187

1220

0.8 ms

0.8 ms

3.1 ms

2.7 ms

GitHub/Official YOLOv7

DFP

218

YOLO-v7-tiny

640×640×3

5.9 M

ONNX

271

273

3.7 ms

3.7 ms

8.2 ms

7.5 ms

GitHub/Official YOLOv7

DFP

219

YOLO-v7-tiny

480×640×3

5.9 M

ONNX

387

404

2.6 ms

2.5 ms

6.5 ms

5.8 ms

GitHub/PINTO

DFP

220

YOLO-v7-tiny

416×416×3

5.9 M

ONNX

653

691

1.5 ms

1.4 ms

4.3 ms

3.8 ms

GitHub/Official YOLOv7

DFP

221

YOLO-v8-medium

640×640×3

24.7 M

ONNX

45

64

22.2 ms

15.6 ms

37.9 ms

27.0 ms

Ultralytics

DFP

222

YOLO-v8-nano

640×640×3

3.0 M

ONNX

327

389

3.1 ms

2.6 ms

6.7 ms

5.8 ms

Ultralytics

DFP

223

YOLO-v8-nano-Oriented-Bounding-Boxes

1024×1024×3

2.9 M

ONNX

128

152

7.8 ms

6.6 ms

13.5 ms

11.7 ms

Ultralytics

DFP

224

YOLO-v8-small

640×640×3

10.6 M

ONNX

186

263

5.4 ms

3.8 ms

10.9 ms

8.0 ms

Ultralytics

DFP

225

YOLO-v8-small-Oriented-Bounding-Boxes

1024×1024×3

10.9 M

ONNX

77

107

13.0 ms

9.3 ms

20.1 ms

15.2 ms

Ultralytics

DFP

226

YOLO-v9-medium

640×640×3

19.1 M

ONNX

40

56

25.0 ms

17.9 ms

61.5 ms

43.7 ms

Ultralytics

DFP

227

YOLO-v9-small

640×640×3

6.9 M

ONNX

109

151

9.2 ms

6.6 ms

32.2 ms

23.0 ms

Ultralytics

DFP

228

YOLO-v9-tiny

640×640×3

2.0 M

ONNX

261

370

3.8 ms

2.7 ms

16.1 ms

11.9 ms

Ultralytics

DFP

229

YOLO-v10-medium

320×320×3

14.6 M

ONNX

166

233

6.0 ms

4.3 ms

18.0 ms

12.9 ms

Ultralytics

DFP

230

YOLO-v10-medium

480×480×3

14.6 M

ONNX

43

60

23.3 ms

16.7 ms

38.6 ms

27.4 ms

Ultralytics

DFP

231

YOLO-v10-nano

320×320×3

2.2 M

ONNX

685

955

1.5 ms

1.0 ms

4.5 ms

3.5 ms

Ultralytics

DFP

232

YOLO-v10-nano

480×480×3

2.2 M

ONNX

121

171

8.3 ms

5.8 ms

12.7 ms

9.6 ms

Ultralytics

DFP

233

YOLO-v10-small

480×480×3

6.9 M

ONNX

106

150

9.4 ms

6.7 ms

16.5 ms

12.2 ms

Ultralytics

DFP

234

YOLO-v10-small

320×320×3

6.9 M

ONNX

319

439

3.1 ms

2.3 ms

8.1 ms

6.0 ms

Ultralytics

DFP

235

YOLO-v11-medium

320×320×3

19.2 M

ONNX

200

283

5.0 ms

3.5 ms

20.1 ms

14.3 ms

Ultralytics

DFP

236

YOLO-v11-nano (MXA-Optimized)

640×640×3

3.9 M

ONNX

326

389

3.1 ms

2.6 ms

8.1 ms

6.6 ms

Ultralytics

DFP

237

YOLO-v11-nano (MXA-Optimized)

480×480×3

3.9 M

ONNX

557

665

1.8 ms

1.5 ms

5.5 ms

4.4 ms

Ultralytics

DFP

238

YOLO-v11-nano (MXA-Optimized)

800×800×3

3.9 M

ONNX

211

251

4.7 ms

4.0 ms

11.0 ms

8.8 ms

Ultralytics

DFP

239

YOLO-v11-nano

480×480×3

2.5 M

ONNX

122

173

8.2 ms

5.8 ms

12.9 ms

9.9 ms

Ultralytics

DFP

240

YOLO-v11-nano

320×320×3

2.5 M

ONNX

776

1088

1.3 ms

0.9 ms

4.5 ms

3.5 ms

Ultralytics

DFP

241

YOLO-v11-small (MXA-Optimized)

640×640×3

14.1 M

ONNX

99

140

10.1 ms

7.1 ms

22.4 ms

16.1 ms

Ultralytics

DFP

242

YOLO-v11-small (MXA-Optimized)

480×480×3

14.1 M

ONNX

220

312

4.5 ms

3.2 ms

15.4 ms

11.0 ms

Ultralytics

DFP

243

YOLO-v11-small

480×480×3

9.0 M

ONNX

102

144

9.8 ms

6.9 ms

17.7 ms

12.9 ms

Ultralytics

DFP

244

YOLO-v11-small

320×320×3

9.0 M

ONNX

316

448

3.2 ms

2.2 ms

8.6 ms

6.3 ms

Ultralytics

DFP

245

YOLOX-medium

640×640×3

24.1 M

ONNX

65

92

15.4 ms

10.9 ms

35.9 ms

25.5 ms

GitHub/YOLOX

DFP

246

YOLOX-medium-ti-lite

416×416×3

24.1 M

ONNX

158

224

6.3 ms

4.5 ms

19.3 ms

13.8 ms

GitHub/YOLOX

DFP

247

YOLOX-medium-ti-lite

640×640×3

24.1 M

ONNX

70

100

14.3 ms

10.0 ms

35.1 ms

25.0 ms

GitHub/YOLOX

DFP

248

YOLOX-nano

416×416×3

883.3 K

ONNX

809

838

1.2 ms

1.2 ms

3.2 ms

2.9 ms

GitHub/YOLOX

DFP

249

YOLOX-nano-ti-lite

416×416×3

2.1 M

ONNX

787

952

1.3 ms

1.1 ms

3.1 ms

2.7 ms

GitHub/YOLOX

DFP

250

YOLOX-small

640×640×3

8.5 M

ONNX

262

371

3.8 ms

2.7 ms

9.4 ms

7.2 ms

OpenMMLab

DFP

251

YOLOX-small-ti-lite

416×416×3

8.5 M

ONNX

510

722

2.0 ms

1.4 ms

5.3 ms

4.0 ms

GitHub/YOLOX

DFP

252

YOLOX-small-ti-lite

640×640×3

8.5 M

ONNX

236

321

4.2 ms

3.1 ms

9.4 ms

7.6 ms

GitHub/YOLOX

DFP

253

YOLOX-tiny

416×416×3

4.8 M

ONNX

644

837

1.6 ms

1.2 ms

4.3 ms

3.6 ms

OpenMMLab

DFP

254

YOLOX-tiny-ti-lite

416×416×3

4.8 M

ONNX

785

953

1.3 ms

1.0 ms

4.4 ms

3.4 ms

GitHub/YOLOX

DFP
Segmentation

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

255

Autoseg-EdgeTPU-small

512×512×3

2.4 M

TFLite

46

66

21.7 ms

15.2 ms

39.5 ms

28.1 ms

Kaggle

DFP

256

BiseNet (ResNet-18)

512×512×3

12.7 M

ONNX

155

220

6.5 ms

4.5 ms

13.7 ms

10.4 ms

OpenMMLab

DFP

257

BiseNet (ResNet-18) - CityScapes

512×512×3

12.7 M

ONNX

144

204

6.9 ms

4.9 ms

15.0 ms

10.8 ms

OpenMMLab

DFP

258

BiseNetv2 (FCN) - CityScapes

512×512×3

3.2 M

ONNX

123

174

8.1 ms

5.7 ms

14.8 ms

10.7 ms

OpenMMLab

DFP

259

Deeplab-EdgeTPU-extra-small

512×512×3

3.1 M

TFLite

410

580

2.4 ms

1.7 ms

5.6 ms

4.4 ms

Kaggle

DFP

260

Deeplab-EdgeTPU-small

512×512×3

4.5 M

TFLite

320

453

3.1 ms

2.2 ms

6.7 ms

4.9 ms

Kaggle

DFP

261

Deeplab-v3

257×257×3

672.2 K

TFLite

1408

1993

0.7 ms

0.5 ms

2.7 ms

2.1 ms

Kaggle

DFP

262

DeepLab-v3 (MobileNet-v2)

513×513×3

672.2 K

TFLite

476

581

2.1 ms

1.7 ms

4.0 ms

3.2 ms

Coral

DFP

263

Deeplab-v3-plus

1024×1024×3

2.1 M

TFLite

27

31

37.0 ms

32.3 ms

47.2 ms

40.8 ms

AI Benchmark

DFP

264

EdgeTPU-DeepLab-slim

513×513×3

2.0 M

TFLite

475

580

2.1 ms

1.7 ms

5.8 ms

4.7 ms

Coral

DFP

265

Edgetpu-segmentation-EdgeTPU

512×512×3

2.2 M

TFLite

46

66

21.7 ms

15.2 ms

35.9 ms

25.6 ms

GitHub/PINTO

DFP

266

Edgetpu-segmentation-EdgeTPU

16×320×16

49.5 K

TFLite

2415

2607

0.4 ms

0.4 ms

1.0 ms

1.0 ms

GitHub/PINTO

DFP

267

ErfNet

208×976×3

2.0 M

ONNX

151

214

6.6 ms

4.7 ms

23.1 ms

16.6 ms

OpenVINO

DFP

268

ERFNet

256×512×3

2.0 M

ONNX

222

260

4.5 ms

3.8 ms

16.2 ms

12.8 ms

GitHub/PINTO

DFP

269

Fast-scnnk-CityScapes

512×512×3

1.3 M

ONNX

476

580

2.1 ms

1.7 ms

4.3 ms

3.4 ms

OpenMMLab

DFP

270

FastSAM-small

640×640×3

11.2 M

ONNX

79

113

12.7 ms

8.8 ms

19.2 ms

14.2 ms

Ultralytics

DFP

271

FPN (ResNet-50) - ADE20K

512×512×3

27.2 M

ONNX

18

26

55.6 ms

38.5 ms

70.4 ms

50.3 ms

OpenMMLab

DFP

272

HRNet

256×192×3

27.2 M

ONNX

115

164

8.7 ms

6.1 ms

45.7 ms

32.5 ms

GitHub/PINTO

DFP

273

Human-segmentation

192×192×3

1.5 M

TFLite

627

866

1.6 ms

1.2 ms

3.3 ms

2.5 ms

GitHub/PINTO

DFP

274

MagicTouch

512×512×4

2.9 M

TFLite

244

345

4.1 ms

2.9 ms

10.2 ms

8.1 ms

GitHub/PINTO

DFP

275

MediaPipe-selfie-segmentation

256×256×3

103.8 K

TFLite

692

935

1.4 ms

1.1 ms

3.9 ms

3.1 ms

MediaPipe

DFP

276

MediaPipe-selfie-segmentation-landscape

144×256×3

103.9 K

TFLite

1415

2002

0.7 ms

0.5 ms

2.6 ms

2.1 ms

MediaPipe

DFP

277

PaNet-ResNet-18-ffm

640×640×3

11.7 M

ONNX

27

38

37.0 ms

26.3 ms

57.8 ms

41.1 ms

OpenMMLab

DFP

278

Selfie-Segmentation

256×256×3

103.9 K

TFLite

766

1071

1.3 ms

0.9 ms

4.5 ms

3.4 ms

GitHub/PINTO

DFP

279

STDC1-CityScapes

512×512×3

7.9 M

ONNX

159

226

6.3 ms

4.4 ms

11.9 ms

8.9 ms

OpenMMLab

DFP

280

STDC2-CityScapes

512×512×3

11.7 M

ONNX

157

223

6.4 ms

4.5 ms

16.8 ms

12.1 ms

OpenMMLab

DFP

281

U-NetMobileNet-v2

256×256×3

6.2 M

TFLite

50

71

20.0 ms

14.1 ms

23.0 ms

16.8 ms

Coral

DFP

282

U-NetMobileNet-v2

128×128×3

6.2 M

TFLite

202

286

5.0 ms

3.5 ms

6.6 ms

5.0 ms

Coral

DFP

283

UNet

1024×1024×3

1.9 M

TFLite

27

32

37.0 ms

31.2 ms

40.5 ms

34.1 ms

AI Benchmark

DFP

284

UNet

256×256×1

29.6 M

ONNX

31

44

32.3 ms

22.7 ms

35.4 ms

25.3 ms

Zetane Viewer

DFP

285

YOLO-v8-medium-seg

640×640×3

26.0 M

ONNX

27

38

37.0 ms

26.3 ms

59.6 ms

42.2 ms

Ultralytics

DFP

286

YOLO-v8-nano-seg

640×640×3

3.2 M

ONNX

207

213

4.8 ms

4.7 ms

9.8 ms

8.8 ms

Ultralytics

DFP

287

YOLO-v11-nano-seg (MXA-Optimized)

640×640×3

4.3 M

ONNX

115

163

8.7 ms

6.1 ms

14.7 ms

11.2 ms

Ultralytics

DFP

288

YOLO-v11-small-seg (MXA-Optimized)

640×640×3

15.3 M

ONNX

38

55

26.3 ms

18.2 ms

39.1 ms

28.1 ms

Ultralytics

DFP
Pose Estimation

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

289

BodyPix (MobileNet)

480×640×3

1.2 M

TFLite

414

497

2.4 ms

2.0 ms

3.9 ms

3.3 ms

Coral

DFP

290

BodyPix (MobileNet)

512×512×3

1.2 M

TFLite

475

576

2.1 ms

1.7 ms

3.5 ms

3.0 ms

Coral

DFP

291

BodyPix (MobileNet)

768×1024×3

1.2 M

TFLite

169

204

5.9 ms

4.9 ms

8.4 ms

7.1 ms

Coral

DFP

292

BodyPix (MobileNet)

576×768×3

1.2 M

TFLite

297

357

3.4 ms

2.8 ms

5.1 ms

4.4 ms

Coral

DFP

293

BodyPix (MobileNet)

720×1280×3

1.2 M

TFLite

146

176

6.8 ms

5.7 ms

9.8 ms

8.2 ms

Coral

DFP

294

BodyPix (MobileNet)

352×480×3

1.2 M

TFLite

805

974

1.2 ms

1.0 ms

2.7 ms

2.2 ms

Coral

DFP

295

BodyPix (ResNet-50)

496×768×3

22.7 M

TFLite

152

215

6.6 ms

4.7 ms

17.4 ms

12.6 ms

Coral

DFP

296

BodyPix (ResNet-50)

624×864×3

22.7 M

TFLite

112

159

8.9 ms

6.3 ms

20.1 ms

14.5 ms

Coral

DFP

297

BodyPix (ResNet-50)

288×416×3

22.7 M

TFLite

247

348

4.0 ms

2.9 ms

12.0 ms

8.8 ms

Coral

DFP

298

BodyPix (ResNet-50)

480×640×3

22.7 M

TFLite

99

139

10.1 ms

7.2 ms

23.0 ms

17.0 ms

Coral

DFP

299

BodyPix (ResNet-50)

672×928×3

22.7 M

TFLite

53

75

18.9 ms

13.3 ms

34.1 ms

24.4 ms

Coral

DFP

300

Handpose-estimation-mediapipe

224×224×3

988.0 K

ONNX

1947

2325

0.5 ms

0.4 ms

1.3 ms

1.1 ms

OpenCV

DFP

301

Handpose-estimation-mediapipe8

224×224×3

904.1 K

ONNX

1477

2091

0.7 ms

0.5 ms

2.8 ms

2.2 ms

OpenCV

DFP

302

HHP-Net

1×5×1, 1×5×1, 1×5×1

36.4 K

TFLite

12142

12142

0.1 ms

0.1 ms

0.2 ms

0.2 ms

GitHub/PINTO

DFP

303

Human-pose-estimation-3d-0001

256×448×3

4.8 M

ONNX

460

652

2.2 ms

1.5 ms

8.6 ms

6.3 ms

OpenVINO

DFP

304

MediaPipe-pose-landk

256×256×3

3.0 M

TFLite

740

876

1.4 ms

1.1 ms

4.4 ms

3.6 ms

MediaPipe

DFP

305

MediaPipe-pose-landk-heavy

256×256×3

13.1 M

TFLite

103

145

9.7 ms

6.9 ms

14.6 ms

10.6 ms

MediaPipe

DFP

306

MediaPipe-pose-landk-lite

256×256×3

1.3 M

TFLite

716

902

1.4 ms

1.1 ms

3.8 ms

3.1 ms

MediaPipe

DFP

307

MoveNet.SinglePose.Lightning

192×192×3

2.2 M

TFLite

2587

2679

0.4 ms

0.4 ms

2.0 ms

1.7 ms

Coral

DFP

308

MoveNet.SinglePose.Thunder

256×256×3

5.9 M

TFLite

1417

1450

0.7 ms

0.7 ms

3.5 ms

3.0 ms

Coral

DFP

309

Pose-estimation-mediapipe

258×258×3

1.3 M

ONNX

567

774

1.8 ms

1.3 ms

4.2 ms

3.4 ms

OpenCV

DFP

310

PoseC3D

48×64×64×17

2.0 M

ONNX

29

41

34.5 ms

24.4 ms

40.9 ms

29.6 ms

GitHub/PINTO

DFP

311

PoseC3D

48×56×56×17

2.9 M

ONNX

33

47

30.3 ms

21.3 ms

35.9 ms

25.7 ms

GitHub/PINTO

DFP

312

PoseNet (MobileNet)

353×481×3

1.2 M

TFLite

803

972

1.2 ms

1.0 ms

2.4 ms

2.0 ms

Coral

DFP

313

PoseNet (MobileNet)

481×641×3

1.2 M

TFLite

414

503

2.4 ms

2.0 ms

4.0 ms

3.3 ms

Coral

DFP

314

PoseNet (MobileNet)

721×1281×3

1.2 M

TFLite

146

177

6.8 ms

5.6 ms

9.3 ms

7.7 ms

Coral

DFP

315

PoseNet (ResNet-50)

288×416×3

22.6 M

TFLite

257

365

3.9 ms

2.7 ms

11.9 ms

8.7 ms

Coral

DFP

316

PoseNet (ResNet-50)

496×768×3

22.6 M

TFLite

152

216

6.6 ms

4.6 ms

16.2 ms

11.7 ms

Coral

DFP

317

PoseNet (ResNet-50)

736×960×3

22.6 M

TFLite

82

117

12.2 ms

8.5 ms

24.2 ms

17.4 ms

Coral

DFP

318

PoseNet (ResNet-50)

624×864×3

22.6 M

TFLite

112

158

8.9 ms

6.3 ms

19.9 ms

14.4 ms

Coral

DFP

319

PoseNet (ResNet-50)

480×640×3

22.6 M

TFLite

116

165

8.6 ms

6.1 ms

19.9 ms

14.4 ms

Coral

DFP

320

PoseNet (ResNet-50)

672×928×3

22.6 M

TFLite

53

75

18.9 ms

13.3 ms

34.5 ms

24.7 ms

Coral

DFP

321

RtmDet-nano-person

320×320×3

962.8 K

ONNX

1199

1443

0.8 ms

0.7 ms

3.0 ms

2.4 ms

OpenMMLab

DFP

322

Simcc-MobileNet-v2-wo-onv

256×192×3

2.2 M

ONNX

1945

2316

0.5 ms

0.4 ms

1.6 ms

1.3 ms

OpenMMLab

DFP

323

YOLO-v8-medium-pose

640×640×3

25.2 M

ONNX

43

61

23.3 ms

16.4 ms

38.3 ms

27.4 ms

Ultralytics

DFP

324

YOLO-v8-nano-pose

640×640×3

3.1 M

ONNX

327

390

3.1 ms

2.6 ms

6.7 ms

5.9 ms

Ultralytics

DFP

325

YOLO-v8-small-pose

640×640×3

11.1 M

ONNX

228

323

4.4 ms

3.1 ms

11.1 ms

8.4 ms

Ultralytics

DFP

326

YOLO-v11-nano-pose (MXA-Optimized)

640×640×3

4.2 M

ONNX

325

390

3.1 ms

2.6 ms

8.2 ms

6.5 ms

Ultralytics

DFP

327

YOLO-v11-small-pose (MXA-Optimized)

640×640×3

14.7 M

ONNX

101

136

9.9 ms

7.4 ms

23.6 ms

17.0 ms

Ultralytics

DFP
Image Enhancement/Super Resolution

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

328

AOD-Net

720×1280×3

1.7 K

TFLite

17

24

58.8 ms

41.7 ms

60.0 ms

42.8 ms

GitHub/PINTO

DFP

329

AOD-Net

480×640×3

1.7 K

TFLite

51

73

19.6 ms

13.7 ms

20.4 ms

14.6 ms

GitHub/PINTO

DFP

330

Colorization-v2

256×256×1

30.7 M

ONNX

31

44

32.3 ms

22.7 ms

50.6 ms

36.1 ms

OpenVINO

DFP

331

DPED

384×512×48

414.7 K

TFLite

41

45

24.4 ms

22.2 ms

36.9 ms

34.6 ms

AI Benchmark

DFP

332

ESRGAN

128×128×3

1.2 M

TFLite

61

86

16.4 ms

11.6 ms

23.8 ms

16.9 ms

AI Benchmark

DFP

333

Fast-SRGAN

512×512×3

154.9 K

TFLite

3

5

333.3 ms

200.0 ms

265.5 ms

187.8 ms

GitHub/PINTO

DFP

334

Fast-SRGAN

128×128×3

154.9 K

TFLite

61

87

16.4 ms

11.5 ms

17.3 ms

12.4 ms

GitHub/PINTO

DFP

335

IMDN

480×640×3

873.0 K

TFLite

8

12

125.0 ms

83.3 ms

129.2 ms

94.3 ms

GitHub/PINTO

DFP

336

IMDN

360×640×3

873.0 K

TFLite

11

16

90.9 ms

62.5 ms

99.4 ms

72.4 ms

GitHub/PINTO

DFP

337

IMDN

1024×1024×3

149.1 K

TFLite

39

41

25.6 ms

24.4 ms

27.5 ms

26.1 ms

AI Benchmark

DFP

338

MBLLEN

480×640×3

439.6 K

TFLite

10

15

100.0 ms

66.7 ms

98.8 ms

70.0 ms

GitHub/PINTO

DFP

339

MBLLEN

720×1280×3

439.6 K

TFLite

3

5

333.3 ms

200.0 ms

289.3 ms

204.8 ms

GitHub/PINTO

DFP

340

Noise2Noise

256×256×3

1.2 M

TFLite

60

86

16.7 ms

11.6 ms

20.8 ms

15.0 ms

GitHub/PINTO

DFP

341

PMN

416×416×4

7.4 M

ONNX

43

61

23.3 ms

16.4 ms

26.5 ms

19.0 ms

GitHub/PINTO

DFP

342

Real-Time-Super-Resolution

240×320×3

1.3 M

TFLite

6

9

166.7 ms

111.1 ms

165.6 ms

117.4 ms

GitHub/PINTO

DFP

343

RFDN

160×160×3

423.3 K

ONNX

22

31

45.5 ms

32.3 ms

75.8 ms

53.9 ms

GitHub/PINTO

DFP

344

RFDN

96×96×3

423.3 K

ONNX

12

17

83.3 ms

58.8 ms

79.5 ms

56.4 ms

GitHub/PINTO

DFP

345

SCI

288×800×3

0.2 K

ONNX

426

557

2.3 ms

1.8 ms

3.2 ms

2.7 ms

GitHub/PINTO

DFP

346

SCL-LLE

240×320×3

77.6 K

TFLite

104

147

9.6 ms

6.8 ms

11.3 ms

8.5 ms

GitHub/PINTO

DFP

347

SCL-LLE

180×320×3

77.6 K

TFLite

138

195

7.2 ms

5.1 ms

8.7 ms

6.5 ms

GitHub/PINTO

DFP

348

SG-Low-Light-Image-Enhancement

24×54×3

10.3 K

ONNX

6654

6971

0.2 ms

0.1 ms

0.4 ms

0.4 ms

GitHub/PINTO

DFP

349

SRCNN-di-v2-k

896×896×3

19.6 K

ONNX

7

10

142.9 ms

100.0 ms

136.8 ms

97.0 ms

OpenMMLab

DFP

350

SRGAN

256×256×3

1.4 M

TFLite

7

11

142.9 ms

90.9 ms

135.1 ms

95.9 ms

AI Benchmark

DFP

351

Super-Resolution

224×224×1

58.3 K

ONNX

160

226

6.2 ms

4.4 ms

6.8 ms

4.9 ms

ONNX Zoo

DFP

352

TBEFN

180×320×3

475.0 K

TFLite

195

205

5.1 ms

4.9 ms

6.6 ms

6.1 ms

GitHub/PINTO

DFP

353

TBEFN

240×320×3

475.0 K

TFLite

146

154

6.8 ms

6.5 ms

8.4 ms

7.8 ms

GitHub/PINTO

DFP

354

VSR

540×960×3

18.6 K

TFLite

29

32

34.5 ms

31.2 ms

51.9 ms

48.0 ms

AI Benchmark

DFP

355

XLSR

360×640×3

21.5 K

TFLite

69

98

14.5 ms

10.2 ms

19.8 ms

15.7 ms

AI Benchmark

DFP

356

Y-Net

480×640×3

190.9 K

TFLite

32

39

31.2 ms

25.6 ms

32.2 ms

26.1 ms

GitHub/PINTO

DFP

357

Y-Net

240×320×3

190.9 K

TFLite

130

160

7.7 ms

6.2 ms

8.6 ms

7.1 ms

GitHub/PINTO

DFP

358

Zero-DCE

360×640×3

77.6 K

TFLite

40

42

25.0 ms

23.8 ms

29.4 ms

28.1 ms

GitHub/PINTO

DFP

359

Zero-DCE

240×320×3

77.6 K

TFLite

121

126

8.3 ms

7.9 ms

10.2 ms

9.6 ms

GitHub/PINTO

DFP

360

Zero-DCE-improved

360×512×3

23.0 K

ONNX

87

123

11.5 ms

8.1 ms

14.8 ms

11.7 ms

GitHub/PINTO

DFP
Face/Head

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

361

ACR-Loss

224×224×3

2.3 M

TFLite

1939

2318

0.5 ms

0.4 ms

1.5 ms

1.2 ms

GitHub/PINTO

DFP

362

BlazeFace

128×128×3

99.0 K

TensorFlow

6836

7891

0.1 ms

0.1 ms

0.5 ms

0.4 ms

GitHub/PINTO

DFP

363

BlazeFace-short-range

128×128×3

99.1 K

TFLite

6835

7898

0.1 ms

0.1 ms

0.5 ms

0.5 ms

GitHub/TorinMB

DFP

364

Face-Detection-0100

256×256×3

1.7 M

TFLite

1682

2017

0.6 ms

0.5 ms

1.6 ms

1.3 ms

GitHub/PINTO

DFP

365

Face-Detection-adas-0001

384×672×3

1.0 M

TFLite

532

650

1.9 ms

1.5 ms

3.2 ms

2.6 ms

GitHub/PINTO

DFP

366

Face-Detection-YuNet

640×640×3

51.9 K

ONNX

328

399

3.0 ms

2.5 ms

4.2 ms

3.6 ms

OpenCV

DFP

367

Face-landk

160×160×3

611.3 K

ONNX

4288

4911

0.2 ms

0.2 ms

0.7 ms

0.6 ms

GitHub/PINTO

DFP

368

Face-Mask-Detection

300×300×3

2.5 M

TFLite

840

1189

1.2 ms

0.8 ms

2.7 ms

2.2 ms

GitHub/PINTO

DFP

369

Face-recognition-sface

112×112×3

9.2 M

ONNX

2580

3652

0.4 ms

0.3 ms

1.4 ms

1.1 ms

OpenCV

DFP

370

FaceBoxes

1024×1024×3

982.3 K

ONNX

128

156

7.8 ms

6.4 ms

9.2 ms

7.9 ms

OpenVINO

DFP

371

FaceMesh

192×192×3

590.5 K

TFLite

3598

4234

0.3 ms

0.2 ms

0.8 ms

0.6 ms

GitHub/PINTO

DFP

372

Head-Pose-Estimation

320×320×3

798.8 K

ONNX

1199

1445

0.8 ms

0.7 ms

2.1 ms

1.7 ms

GitHub/YinguoBing

DFP

373

MediaPipe-face-Detection-range

192×192×3

502.1 K

TFLite

3579

4185

0.3 ms

0.2 ms

1.4 ms

1.2 ms

MediaPipe

DFP

374

MediaPipe-face-Detection-short-range

128×128×3

99.1 K

TFLite

6836

7888

0.1 ms

0.1 ms

0.5 ms

0.5 ms

MediaPipe

DFP

375

MediaPipe-face-landk

192×192×3

590.5 K

TFLite

3597

4231

0.3 ms

0.2 ms

0.8 ms

0.6 ms

MediaPipe

DFP

376

MediaPipe-iris-landk

64×64×3

636.0 K

TFLite

10633

10633

0.1 ms

0.1 ms

0.6 ms

0.6 ms

MediaPipe

DFP

377

RetinaFacewithResNet-50

640×640×3

26.0 M

ONNX

45

63

22.2 ms

15.9 ms

33.8 ms

24.2 ms

OpenVINO

DFP

378

SFace

112×112×3

9.2 M

TFLite

2580

3653

0.4 ms

0.3 ms

1.4 ms

1.1 ms

GitHub/PINTO

DFP

379

SSD-MobileNet-v2 (Face)

320×320×3

5.1 M

TFLite

1199

1442

0.8 ms

0.7 ms

2.5 ms

2.0 ms

Coral

DFP

380

SynergyNet

120×120×3

2.2 M

ONNX

4943

6989

0.2 ms

0.1 ms

1.0 ms

0.8 ms

GitHub/PINTO

DFP

381

UltraLightweightFaceDetectionRFB320

240×320×3

274.5 K

ONNX

1692

2019

0.6 ms

0.5 ms

1.3 ms

1.1 ms

OpenVINO

DFP

382

Version-RFB

240×320×3

274.5 K

ONNX

1691

2019

0.6 ms

0.5 ms

1.2 ms

1.1 ms

ONNX Zoo

DFP

383

WHENet

224×224×3

4.1 M

TFLite

970

1374

1.0 ms

0.7 ms

6.4 ms

4.8 ms

GitHub/PINTO

DFP

384

YOLO-v5-Face

256×320×3

1.6 M

TensorFlow

1638

1990

0.6 ms

0.5 ms

2.7 ms

2.1 ms

GitHub/PINTO

DFP

385

YuNet

120×160×3

81.1 K

TensorFlow

5942

6876

0.2 ms

0.1 ms

0.5 ms

0.4 ms

GitHub/PINTO

DFP

386

YuNet-v2

640×640×3

51.9 K

TFLite

328

399

3.0 ms

2.5 ms

4.2 ms

3.6 ms

GitHub/PINTO

DFP
Classification (Other)

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

387

Aiy-vision-classifier-birds

224×224×3

3.3 M

TFLite

1785

2313

0.6 ms

0.4 ms

1.9 ms

1.5 ms

Kaggle

DFP

388

Aiy-vision-classifier-food

192×192×3

5.0 M

TFLite

3584

4225

0.3 ms

0.2 ms

1.3 ms

1.0 ms

Kaggle

DFP

389

Aiy-vision-classifier-insects

224×224×3

3.4 M

TFLite

1785

2311

0.6 ms

0.4 ms

1.9 ms

1.5 ms

Kaggle

DFP

390

Aiy-vision-classifier-plants

224×224×3

4.7 M

TFLite

1785

2318

0.6 ms

0.4 ms

2.0 ms

1.5 ms

Kaggle

DFP

391

Anti-spoof-mn3

128×128×3

2.9 M

TFLite

3883

5491

0.3 ms

0.2 ms

1.7 ms

1.2 ms

GitHub/PINTO

DFP

392

CropNet-classifier-cassava-disease

224×224×3

4.0 M

TFLite

1946

2311

0.5 ms

0.4 ms

3.0 ms

2.3 ms

Kaggle

DFP

393

Gender-googleNet

224×224×3

5.7 M

ONNX

518

734

1.9 ms

1.4 ms

3.7 ms

2.8 ms

ONNX Zoo

DFP

394

Malaria-Detector

125×125×3

6.7 M

Keras

6855

8108

0.1 ms

0.1 ms

0.5 ms

0.4 ms

xmartlabs

DFP

395

NSFW

224×224×3

5.6 M

TFLite

1834

2235

0.5 ms

0.4 ms

1.9 ms

1.5 ms

GitHub/PINTO

DFP

396

Ocr-japanese

512×512×1

23.1 M

TensorFlow

14

20

71.4 ms

50.0 ms

94.9 ms

67.2 ms

GitHub/PINTO

DFP

397

Person-Attributes-recog-crossroad-0230

160×80×3

717.7 K

TFLite

5825

8200

0.2 ms

0.1 ms

0.7 ms

0.6 ms

GitHub/PINTO

DFP

398

Vehicle-Attributes-recog-barrier-0039

72×72×3

611.5 K

TFLite

7133

8793

0.1 ms

0.1 ms

0.6 ms

0.5 ms

GitHub/PINTO

DFP

399

Vehicle-Attributes-recog-barrier-0042

72×72×3

10.7 M

TFLite

5911

8362

0.2 ms

0.1 ms

1.2 ms

1.0 ms

GitHub/PINTO

DFP
ADAS/Driver Monitoring

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

400

Driver-action-recognition-adas-0002-encod…

224×224×3

2.7 M

TFLite

1948

2312

0.5 ms

0.4 ms

1.6 ms

1.3 ms

GitHub/PINTO

DFP

401

Gaze-estimation-adas-0002

60×60×3, 60×60×3, 3

1.8 M

TFLite

268

268

3.7 ms

3.7 ms

4.4 ms

4.3 ms

GitHub/PINTO

DFP

402

Head-pose-estimation-adas-0001

60×60×3

1.8 M

TFLite

8399

11650

0.1 ms

0.1 ms

0.7 ms

0.6 ms

GitHub/PINTO

DFP

403

Open-closed-eye-0001

32×32×3

11.0 K

ONNX

19501

19646

0.1 ms

0.1 ms

0.2 ms

0.2 ms

OpenVINO

DFP

404

Pedestrian-and-vehicle-Detector-adas-0001

384×672×3

1.6 M

TFLite

531

645

1.9 ms

1.6 ms

3.7 ms

3.1 ms

GitHub/PINTO

DFP

405

Pedestrian-Detection-adas-0002

384×672×3

1.1 M

TFLite

532

648

1.9 ms

1.5 ms

3.4 ms

2.9 ms

GitHub/PINTO

DFP

406

Road-segmentation-adas-0001

512×896×3

180.0 K

TFLite

145

186

6.9 ms

5.4 ms

9.6 ms

7.3 ms

GitHub/PINTO

DFP

407

Vehicle-Detection-0200

256×256×3

1.7 M

TFLite

1688

2019

0.6 ms

0.5 ms

1.6 ms

1.3 ms

GitHub/PINTO

DFP

408

YOLOp320

320×320×3

7.6 M

ONNX

402

570

2.5 ms

1.8 ms

7.1 ms

5.2 ms

GitHub/YOLOP

DFP

409

YOLOp640

640×640×3

7.6 M

ONNX

134

190

7.5 ms

5.3 ms

17.6 ms

12.7 ms

GitHub/YOLOP

DFP
Hand/Palm

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

410

Hand-Detection-and-Tracking

256×256×3

1.3 M

TensorFlow

1034

1126

1.0 ms

0.9 ms

4.2 ms

3.3 ms

GitHub/PINTO

DFP

411

Hand-recrop

256×256×3

26.1 K

TFLite

1693

2029

0.6 ms

0.5 ms

0.9 ms

0.8 ms

GitHub/PINTO

DFP

412

MediaPipe-hand-landk

224×224×3

2.6 M

TFLite

1951

2318

0.5 ms

0.4 ms

1.8 ms

1.4 ms

MediaPipe

DFP

413

MediaPipe-hand-landk-lite

224×224×3

988.0 K

TFLite

1954

2322

0.5 ms

0.4 ms

1.4 ms

1.1 ms

MediaPipe

DFP

414

MediaPipe-hand-recrop

256×256×3

26.4 K

TFLite

1691

2030

0.6 ms

0.5 ms

0.9 ms

0.8 ms

MediaPipe

DFP

415

MediaPipe-palm-Detection

192×192×3

1.1 M

TFLite

907

1282

1.1 ms

0.8 ms

2.6 ms

2.0 ms

MediaPipe

DFP

416

MediaPipe-palm-Detection-lite

192×192×3

937.6 K

TFLite

1259

1737

0.8 ms

0.6 ms

2.1 ms

1.6 ms

MediaPipe

DFP

417

Palm-Detection-mediapipe

192×192×3

937.6 K

ONNX

1144

1554

0.9 ms

0.6 ms

2.2 ms

1.7 ms

OpenCV

DFP
Depth Estimation

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

418

Graft-PSMNet

256×120×160

1.2 M

ONNX

96

96

10.4 ms

10.4 ms

15.1 ms

15.1 ms

GitHub/PINTO

DFP

419

Graft-PSMNet

256×80×120

1.2 M

ONNX

193

193

5.2 ms

5.2 ms

8.2 ms

8.2 ms

GitHub/PINTO

DFP

420

HR-Depth

384×1280×3

10.7 M

ONNX

53

54

18.9 ms

18.5 ms

25.5 ms

25.3 ms

GitHub/PINTO

DFP

421

MiDaS

256×256×3

15.8 M

TFLite

216

305

4.6 ms

3.3 ms

14.3 ms

10.3 ms

Kaggle

DFP

422

PyDNet

192×320×3

1.9 M

TFLite

354

501

2.8 ms

2.0 ms

5.0 ms

3.7 ms

GitHub/PINTO

DFP

423

PyDNet2

384×1280×3

699.9 K

TFLite

44

63

22.7 ms

15.9 ms

27.1 ms

19.9 ms

GitHub/PINTO

DFP

424

PyDNet2

384×640×3

699.9 K

TFLite

88

125

11.4 ms

8.0 ms

14.0 ms

10.3 ms

GitHub/PINTO

DFP
Point Cloud

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

425

MMDet3d-CenterPoint

800×800×64

4.9 M

ONNX

11

11

90.9 ms

90.9 ms

112.7 ms

112.7 ms

GitHub/Tartisan

DFP

426

MMDet3d-CenterPoint

20×10

0.6 K

ONNX

38192

38192

0.0 ms

0.0 ms

0.2 ms

0.2 ms

GitHub/Tartisan

DFP

427

MMDet3d-PointPillars

468×468×64

4.6 M

ONNX

2

3

500.0 ms

333.3 ms

409.5 ms

296.0 ms

GitHub/Tartisan

DFP

428

MMDet3d-PointPillars

20×10

0.6 K

ONNX

37963

38114

0.0 ms

0.0 ms

0.2 ms

0.2 ms

GitHub/Tartisan

DFP

429

PointPillars (SECFPN) - KITTI 3D

496×432×64

4.6 M

ONNX

10

14

100.0 ms

71.4 ms

110.8 ms

82.4 ms

OpenMMLab

DFP

430

PointPillars (SECFPN-SBN) - NUS 3D

400×400×64

4.7 M

ONNX

13

18

76.9 ms

55.6 ms

94.3 ms

72.2 ms

OpenMMLab

DFP

431

SFA3D

608×608×3

12.1 M

TensorFlow

89

95

11.2 ms

10.5 ms

19.3 ms

16.6 ms

GitHub/PINTO

DFP
Audio/Speech

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

432

Audio-classification (YamNet)

96×64×1

3.6 M

TFLite

11571

12028

0.1 ms

0.1 ms

0.8 ms

0.7 ms

Kaggle

DFP

433

Audio-Denoising (UNet)

128×128×1

1.9 M

Keras

1736

2117

0.6 ms

0.5 ms

1.8 ms

1.3 ms

GitHub/vbelz

DFP

434

FRILL

96×64×1

9.6 M

TFLite

5465

7730

0.2 ms

0.1 ms

1.3 ms

1.0 ms

GitHub/PINTO

DFP

435

Light-SERNet

184×40×1

449.9 K

TFLite

3221

3365

0.3 ms

0.3 ms

0.7 ms

0.6 ms

GitHub/PINTO

DFP

436

Speech-enhancement

128×128×1

1.9 M

ONNX

967

1370

1.0 ms

0.7 ms

2.1 ms

1.6 ms

GitHub/PINTO

DFP

437

YAMNet

96×64×1

3.6 M

TFLite

11596

11994

0.1 ms

0.1 ms

0.8 ms

0.7 ms

GitHub/PINTO

DFP
Emotion Detection

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

438

Emotion-FERPlus

64×64×1

8.4 M

TFLite

965

1367

1.0 ms

0.7 ms

1.7 ms

1.3 ms

GitHub/PINTO

DFP

439

Emotion-ferplus

64×64×1

8.4 M

ONNX

991

1404

1.0 ms

0.7 ms

1.7 ms

1.3 ms

ONNX Zoo

DFP

440

Facial-expRession-recognition-MobilefaceN…

112×112×3

1.1 M

TFLite

5725

7624

0.2 ms

0.1 ms

0.9 ms

0.8 ms

GitHub/PINTO

DFP

441

Facial-expRession-recognition-MobilefaceN…

112×112×3

1.1 M

ONNX

5722

7622

0.2 ms

0.1 ms

0.9 ms

0.8 ms

OpenCV

DFP

442

Facial-expRession-recognition-MobilefaceN…

112×112×3

1.1 M

ONNX

978

1386

1.0 ms

0.7 ms

5.4 ms

3.9 ms

OpenCV

DFP
Style Transfer/GANs

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

443

Arbitrary-stylization

256×256×3

2.2 M

TFLite

1691

2024

0.6 ms

0.5 ms

1.6 ms

1.3 ms

GitHub/PINTO

DFP

444

Arbitrary-stylization-inception-v3

256×256×3

8.6 M

TFLite

566

802

1.8 ms

1.2 ms

4.7 ms

3.5 ms

Kaggle

DFP

445

Facial-cartoonizer

512×512×3

2.0 M

ONNX

61

87

16.4 ms

11.5 ms

19.3 ms

13.8 ms

GitHub/Facial Image Cartoonizaiton

DFP

446

MobileStyleGAN

512

2.0 M

ONNX

37631

37631

0.0 ms

0.0 ms

0.2 ms

0.2 ms

GitHub/PINTO

DFP
Gesture Recognition

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

447

American-sign-language

224×224×3

2.7 M

TFLite

1946

2314

0.5 ms

0.4 ms

1.5 ms

1.2 ms

Kaggle

DFP

448

FINNger

96×96×3

3.2 M

TensorFlow

378

536

2.6 ms

1.9 ms

3.8 ms

2.8 ms

GitHub/PINTO

DFP

449

Naruto-handsign-Detection

416×416×3

871.1 K

TFLite

809

846

1.2 ms

1.2 ms

3.1 ms

2.8 ms

GitHub/PINTO

DFP
Tracking

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

450

LiteTrack

127×127×3

1.2 M

TFLite

2447

3462

0.4 ms

0.3 ms

2.6 ms

2.0 ms

GitHub/PINTO

DFP

451

OpenCV-DaSiamRPN (Feature Extractor)

360×480×3

4.8 M

ONNX

93

132

10.8 ms

7.6 ms

12.4 ms

8.9 ms

OpenCV

DFP

452

OpenCV-DaSiamRPN (Kernel-R1)

256×24×24

11.3 M

ONNX

110

114

9.1 ms

8.8 ms

11.3 ms

10.7 ms

OpenCV

DFP
Wireframe/keypoint Detection

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

453

M-LSD

320×320×4

599.8 K

TFLite

804

1095

1.2 ms

0.9 ms

2.7 ms

2.2 ms

GitHub/PINTO

DFP

454

M-LSD

512×512×4

1.5 M

TFLite

31

44

32.3 ms

22.7 ms

35.9 ms

26.1 ms

GitHub/PINTO

DFP
Re-identification

#

Name

Input Resolution

Params Count

F/W

FPS (14T)

FPS (20T)

Latency 1 (14T)

Latency 1 (20T)

Latency 2 (14T)

Latency 2 (20T)

Source

DFP

455

Person-ReID-YouTube

256×128×3

25.5 M

ONNX

735

1042

1.4 ms

1.0 ms

4.1 ms

3.0 ms

OpenCV

DFP
Third-Party Licenses

Each of these models is shared under the original creator’s license. Please check the license at the source page.