Here, we give the full list of publicly pre-trained models supported by the Hailo Model Zoo.
- Network available in Hailo Benchmark are marked with
- Networks available in TAPPAS are marked with
- Benchmark and TAPPAS networks run in performance mode
- All models were compiled using Hailo Dataflow Compiler v3.29.0
| Network Name | Accuracy (top1) | HW Accuracy | FPS (Batch Size=1) | FPS (Batch Size=8) | Input Resolution (HxWxC) | Params (M) | OPS (G) | Pretrained | Source | Compiled |
|---|---|---|---|---|---|---|---|---|---|---|
| cas_vit_m | 81.2 | 81.01 | 36 | 106 | 384x384x3 | 12.42 | 10.89 | download | link | download |
| cas_vit_s | 79.93 | 79.75 | 52 | 155 | 384x384x3 | 5.5 | 5.4 | download | link | download |
| cas_vit_t | 81.9 | 81.58 | 23 | 63 | 384x384x3 | 21.76 | 20.85 | download | link | download |
| deit_base | 80.93 | 80.19 | 28 | 107 | 224x224x3 | 80.26 | 35.22 | download | link | download |
| deit_small | 76.25 | 75.38 | 38 | 125 | 224x224x3 | 20.52 | 9.4 | download | link | download |
| deit_tiny | 69.07 | 68.62 | 74 | 302 | 224x224x3 | 5.3 | 2.57 | download | link | download |
| efficientnet_l | 80.47 | 79.3 | 221 | 221 | 300x300x3 | 10.55 | 19.4 | download | link | download |
| efficientnet_lite0 | 74.99 | 73.85 | 1731 | 1731 | 224x224x3 | 4.63 | 0.78 | download | link | download |
| efficientnet_lite1 | 76.67 | 76.3 | 934 | 934 | 240x240x3 | 5.39 | 1.22 | download | link | download |
| efficientnet_lite2 | 77.46 | 76.73 | 429 | 429 | 260x260x3 | 6.06 | 1.74 | download | link | download |
| efficientnet_lite3 | 79.29 | 78.66 | 223 | 223 | 280x280x3 | 8.16 | 2.8 | download | link | download |
| efficientnet_lite4 | 80.79 | 80.08 | 85 | 250 | 300x300x3 | 12.95 | 5.10 | download | link | download |
efficientnet_m  |
78.91 | 78.48 | 984 | 984 | 240x240x3 | 6.87 | 7.32 | download | link | download |
| efficientnet_s | 77.63 | 77.3 | 1036 | 1036 | 224x224x3 | 5.41 | 4.72 | download | link | download |
| fastvit_sa12 | 76.8 | 76.49 | 273 | 273 | 224x224x3 | 11.99 | 3.59 | download | link | download |
| hardnet39ds | 73.43 | 73.01 | 333 | 1296 | 224x224x3 | 3.48 | 0.86 | download | link | download |
| hardnet68 | 75.47 | 75.27 | 123 | 358 | 224x224x3 | 17.56 | 8.5 | download | link | download |
| inception_v1 | 69.74 | 69.52 | 928 | 928 | 224x224x3 | 6.62 | 3 | download | link | download |
| mobilenet_v1 | 70.97 | 70.3 | 3100 | 3100 | 224x224x3 | 4.22 | 1.14 | download | link | download |
| mobilenet_v2_1.0 |
71.78 | 70.95 | 2596 | 2596 | 224x224x3 | 3.49 | 0.62 | download | link | download |
| mobilenet_v2_1.4 | 74.18 | 73.18 | 1668 | 1668 | 224x224x3 | 6.09 | 1.18 | download | link | download |
| mobilenet_v3 | 72.21 | 71.8 | 2400 | 2401 | 224x224x3 | 4.07 | 2 | download | link | download |
| mobilenet_v3_large_minimalistic | 72.12 | 70.61 | 3005 | 3005 | 224x224x3 | 3.91 | 0.42 | download | link | download |
| regnetx_1.6gf | 77.05 | 76.71 | 2321 | 2321 | 224x224x3 | 9.17 | 3.22 | download | link | download |
| regnetx_800mf | 75.16 | 74.83 | 3505 | 3505 | 224x224x3 | 7.24 | 1.6 | download | link | download |
| repghost_1_0x | 73.03 | 72.28 | 205 | 861 | 224x224x3 | 4.1 | 0.28 | download | link | download |
| repghost_2_0x | 77.18 | 76.91 | 115 | 480 | 224x224x3 | 9.8 | 1.04 | download | link | download |
| repvgg_a1 | 74.4 | 72.35 | 2545 | 2545 | 224x224x3 | 12.79 | 4.7 | download | link | download |
| repvgg_a2 | 76.52 | 74.48 | 911 | 911 | 224x224x3 | 25.5 | 10.2 | download | link | download |
| resmlp12_relu | 75.27 | 74.87 | 1429 | 1429 | 224x224x3 | 15.77 | 6.04 | download | link | download |
| resnet_v1_18 | 71.27 | 71.09 | 2533 | 2533 | 224x224x3 | 11.68 | 3.64 | download | link | download |
| resnet_v1_34 | 72.7 | 72.27 | 1346 | 1346 | 224x224x3 | 21.79 | 7.34 | download | link | download |
resnet_v1_50  |
75.21 | 74.67 | 1371 | 1371 | 224x224x3 | 25.53 | 6.98 | download | link | download |
| resnext26_32x4d | 76.17 | 75.94 | 1630 | 1630 | 224x224x3 | 15.37 | 4.96 | download | link | download |
| resnext50_32x4d | 79.3 | 78.4 | 414 | 414 | 224x224x3 | 24.99 | 8.48 | download | link | download |
| squeezenet_v1.1 | 59.85 | 59.39 | 3034 | 3034 | 224x224x3 | 1.24 | 0.78 | download | link | download |
| vit_base | 84.5 | 83.16 | 23 | 72 | 224x224x3 | 86.5 | 35.188 | download | link | download |
| vit_base_bn |
79.98 | 78.82 | 39 | 139 | 224x224x3 | 86.5 | 35.188 | download | link | download |
| vit_small | 81.5 | 79.91 | 48 | 187 | 224x224x3 | 21.12 | 8.62 | download | link | download |
| vit_small_bn | 78.12 | 77.24 | 128 | 617 | 224x224x3 | 21.12 | 8.62 | download | link | download |
| vit_tiny | 75.51 | 73.6 | 75 | 303 | 224x224x3 | 5.73 | 2.2 | download | link | download |
| vit_tiny_bn | 68.95 | 67.28 | 191 | 1033 | 224x224x3 | 5.73 | 2.2 | download | link | download |