Here, we give the full list of publicly pre-trained models supported by the Hailo Model Zoo.
- Benchmark Networks 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 | mAP | HW Accuracy | FPS (Batch Size=1) | FPS (Batch Size=8) | Input Resolution (HxWxC) | Params (M) | OPS (G) | Pretrained | Source | Compiled |
|---|---|---|---|---|---|---|---|---|---|---|
| centernet_resnet_v1_18_postprocess | 26.37 | 25.02 | 125 | 201 | 512x512x3 | 14.22 | 31.21 | download | link | rgbx/nv12 |
| centernet_resnet_v1_50_postprocess | 31.77 | 29.27 | 80 | 116 | 512x512x3 | 30.07 | 56.92 | download | link | rgbx/nv12 |
| damoyolo_tinynasL20_T | 42.8 | 41.32 | 144 | 293 | 640x640x3 | 11.35 | 18.02 | download | link | rgbx/nv12 |
| damoyolo_tinynasL25_S | 46.53 | 45.31 | 109 | 229 | 640x640x3 | 16.25 | 37.64 | download | link | rgbx/nv12 |
| damoyolo_tinynasL35_M | 49.7 | 47.97 | 58 | 105 | 640x640x3 | 33.98 | 61.64 | download | link | rgbx/nv12 |
| detr_resnet_v1_18_bn | 33.91 | 31.59 | 31 | 64 | 800x800x3 | 32.42 | 61.87 | download | link | rgbx/nv12 |
| detr_resnet_v1_50 | 35.38 | 34.89 | 12 | 18 | 800x800x3 | 41.1 | 120.4 | download | link | rgbx |
| efficientdet_lite0 | 27.32 | 26.54 | 117 | 294 | 320x320x3 | 3.56 | 1.94 | download | link | rgbx/nv12 |
| efficientdet_lite1 | 32.27 | 31.82 | 79 | 176 | 384x384x3 | 4.73 | 4 | download | link | rgbx/nv12 |
| efficientdet_lite2 | 35.95 | 34.75 | 51 | 99 | 448x448x3 | 5.93 | 6.84 | download | link | rgbx/nv12 |
nanodet_repvgg  |
29.3 | 28.59 | 705 | 707 | 416x416x3 | 6.74 | 11.28 | download | link | rgbx/nv12 |
| nanodet_repvgg_a12 | 33.73 | 31.5 | 156 | 260 | 640x640x3 | 5.13 | 28.23 | download | link | rgbx/nv12 |
| nanodet_repvgg_a1_640 | 33.28 | 32.94 | 181 | 181 | 640x640x3 | 10.79 | 42.8 | download | link | rgbx/nv12 |
ssd_mobilenet_v1  |
23.19 | 22.41 | 1147 | 1147 | 300x300x3 | 6.79 | 2.5 | download | link | rgbx/nv12 |
| ssd_mobilenet_v2 | 24.18 | 23.02 | 175 | 345 | 300x300x3 | 4.46 | 1.52 | download | link | rgbx/nv12 |
| tiny_yolov3 | 14.66 | 14.42 | 1045 | 1045 | 416x416x3 | 8.85 | 5.58 | download | link | rgbx/nv12 |
| tiny_yolov4 | 19.18 | 17.62 | 907 | 907 | 416x416x3 | 6.05 | 6.92 | download | link | rgbx/nv12 |
| yolov10b | 52.0 | 51.16 | 33 | 54 | 640x640x3 | 20.15 | 92.09 | download | link | rgbx/nv12 |
| yolov10n | 38.5 | 37.12 | 182 | 426 | 640x640x3 | 2.3 | 6.8 | download | link | rgbx/nv12 |
| yolov10s | 45.86 | 45.16 | 96 | 208 | 640x640x3 | 7.2 | 21.7 | download | link | rgbx/nv12 |
| yolov10x | 53.7 | 51.92 | 18 | 28 | 640x640x3 | 31.72 | 160.56 | download | link | rgbx/nv12 |
| yolov3 | 38.42 | 38.31 | 34 | 45 | 608x608x3 | 68.79 | 158.10 | download | link | rgbx/nv12 |
| yolov3_416 | 37.73 | 37.52 | 52 | 80 | 416x416x3 | 61.92 | 65.94 | download | link | rgbx/nv12 |
| yolov3_gluon | 37.28 | 35.77 | 36 | 47 | 608x608x3 | 68.79 | 140.7 | download | link | rgbx/nv12 |
| yolov3_gluon_416 | 36.27 | 34.15 | 52 | 80 | 416x416x3 | 61.92 | 65.94 | download | link | rgbx/nv12 |
| yolov4_leaky | 42.37 | 41.08 | 44 | 69 | 512x512x3 | 64.33 | 91.04 | download | link | rgbx/nv12 |
| yolov5m | 42.59 | 41.44 | 78 | 123 | 640x640x3 | 21.78 | 52.17 | download | link | rgbx/nv12 |
| yolov5m6_6.1 | 50.68 | 49.36 | 25 | 32 | 1280x1280x3 | 35.70 | 200.04 | download | link | rgbx/nv12 |
| yolov5m_6.1 | 44.74 | 43.47 | 80 | 125 | 640x640x3 | 21.17 | 48.96 | download | link | rgbx/nv12 |
| yolov5m_wo_spp |
43.06 | 41.56 | 108 | 202 | 640x640x3 | 22.67 | 52.88 | download | link | rgbx/nv12 |
| yolov5s | 35.33 | 34.1 | 162 | 278 | 640x640x3 | 7.46 | 17.44 | download | link | rgbx |
| yolov5s_c3tr | 37.13 | 35.83 | 141 | 285 | 640x640x3 | 10.29 | 17.02 | download | link | rgbx/nv12 |
| yolov5s_wo_spp | 34.79 | 33.81 | 182 | 303 | 640x640x3 | 7.85 | 17.74 | download | link | rgbx/nv12 |
| yolov5xs_wo_spp | 33.18 | 32.26 | 246 | 484 | 512x512x3 | 7.85 | 11.36 | download | link | rgbx/nv12 |
| yolov5xs_wo_spp_nms_core | 32.73 | 31.93 | 246 | 484 | 512x512x3 | 7.85 | 11.36 | download | link | rgbx/nv12 |
| yolov6n | 34.29 | 32.45 | 240 | 492 | 640x640x3 | 4.32 | 11.12 | download | link | rgbx/nv12 |
| yolov6n_0.2.1 | 35.16 | 33.97 | 249 | 519 | 640x640x3 | 4.33 | 11.06 | download | link | rgbx |
| yolov6n_0.2.1_nms_core | 35.16 | 34.02 | 120 | 179 | 640x640x3 | 4.32 | 11.12 | download | link | rgbx |
| yolov7 | 50.6 | 48.9 | 40 | 59 | 640x640x3 | 36.91 | 104.51 | download | link | rgbx/nv12 |
| yolov7_tiny | 37.07 | 36.22 | 182 | 311 | 640x640x3 | 6.22 | 13.74 | download | link | rgbx/nv12 |
| yolov7e6 | 55.37 | 53.21 | 9 | 11 | 1280x1280x3 | 97.20 | 515.12 | download | link | rgbx/nv12 |
| yolov8l | 52.44 | 51.84 | 26 | 39 | 640x640x3 | 43.7 | 165.3 | download | link | rgbx/nv12 |
| yolov8m |
49.91 | 49.32 | 72 | 141 | 640x640x3 | 25.9 | 78.93 | download | link | rgbx/nv12 |
| yolov8n | 37.02 | 36.46 | 289 | 535 | 640x640x3 | 3.2 | 8.74 | download | link | rgbx/nv12 |
| yolov8s | 44.58 | 43.98 | 124 | 212 | 640x640x3 | 11.2 | 28.6 | download | link | rgbx/nv12 |
| yolov8x | 53.45 | 52.8 | 19 | 27 | 640x640x3 | 68.2 | 258 | download | link | rgbx/nv12 |
| yolov9c | 52.6 | 51.29 | 34 | 49 | 640x640x3 | 25.3 | 102.1 | download | link | rgbx/nv12 |
| yolox_l_leaky |
48.68 | 46.48 | 31 | 43 | 640x640x3 | 54.17 | 155.3 | download | link | rgbx/nv12 |
| yolox_s_leaky | 38.13 | 37.29 | 129 | 214 | 640x640x3 | 8.96 | 26.74 | download | link | rgbx/nv12 |
| yolox_s_wide_leaky | 42.0 | 41.0 | 77 | 115 | 640x640x3 | 20.12 | 59.46 | download | link | rgbx/nv12 |
| yolox_tiny | 32.64 | 31.4 | 268 | 583 | 416x416x3 | 5.05 | 6.44 | download | link | rgbx/nv12 |