# YOLOv4 model zoo

We provide a collection of [COCO]() pre-trained files of [YOLOv4]() series models.

## COCO pre-trained files (512x512 random)

Models used anchors of 512x512 input resolution for training.

| Model | Size | AP | AP₅₀ | AP₇₅ | AP_S | AP_M | AP_L | cfg | weights | score |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| **YOLOv4** | 512 | 43.0 | 64.9 | 46.5 | 24.3 | 46.1 | 55.2 | [cfg](https://drive.google.com/open?id=1hSrVqiEbuVewEIVU4cqG-SqJFUNTYmIC) | [weights](https://drive.google.com/open?id=1L-SO373Udc9tPz5yLkgti5IAXFboVhUt) | [score]() |
| **YOLOv4-Leaky** | 512 | 42.4 | 64.5 | 46.0 | 23.9 | 45.6 | 54.2 | [cfg](https://drive.google.com/open?id=1C4w_2loEpi-MznqMgKo16oZD7BcvgsCF) | [weights](https://drive.google.com/open?id=1dW-Sd70aTmXuFvYspiY85SwWdn0cr447) | [score]() |
| **YOLOv4-SAM-Leaky** | 512 | 42.7 | 64.5 | 46.4 | 24.1 | 45.3 | 54.8 | [cfg](https://drive.google.com/open?id=1gLCJI1KGD6mVdqy4K8jlxKHvwk1Texga) | [weights](https://drive.google.com/open?id=1HMJa5OUI4QG33ZkUEcyC8bfWDJUBB4qF) | [score]() |
| **YOLOv4-Mish** | 512 | 43.2 | 64.7 | 46.9 | 24.4 | 46.1 | 55.4 | [cfg](https://drive.google.com/open?id=1theQTAhnmtFQHtrnm7IFHbGHI6KGgKim) | [weights](https://drive.google.com/open?id=1sWNozS0emz7bmQTUWDLvsubLGnCwUiIS) | [score]() |
| **YOLOv4-SAM-Mish** | 512 | 43.4 | 65.0 | 47.2 | 24.6 | **46.7** | 55.5 | [cfg](https://drive.google.com/open?id=1V726wzWW1iBDcJ7uP8c0n_Y4M98lIg-C) | [weights](https://drive.google.com/open?id=1wK66ga9YgtjGNSm9fpouJn2GaDxa7SfT) | [score]() |
| **YOLOv4-CSP** | 512 | 43.6 | 64.9 | 47.2 | 24.3 | 46.5 | 56.5 | [cfg](https://drive.google.com/file/d/1XwtHTwn5mkS-Lqdn80PcULNyK-gmpEEm/view?usp=sharing) | [weights](https://drive.google.com/file/d/1FCGAQsBADMPdyfmZrOsSDeneikGWcnlu/view?usp=sharing) | [score]() |
| **YOLOv4-CSP-SAM** | 512 | **43.7** | **65.2** | **47.4** | **24.7** | **46.7** | **56.6** | [cfg](https://drive.google.com/file/d/1h7amsxwtWz6v66kzx53R3_Sxst5pti06/view?usp=sharing) | [weights](https://drive.google.com/file/d/1xBdS-r-iH44ZdmoXp5z5BlmHezzxNNUC/view?usp=sharing) | [score]() |

| Model | Size | AP | AP₅₀ | AP₇₅ | AP_S | AP_M | AP_L | cfg | weights | score |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| **YOLOv4** | 608 | 43.5 | 65.7 | 47.3 | 26.7 | 46.7 | 53.3 | [cfg](https://drive.google.com/open?id=1hSrVqiEbuVewEIVU4cqG-SqJFUNTYmIC) | [weights](https://drive.google.com/open?id=1L-SO373Udc9tPz5yLkgti5IAXFboVhUt) | [score]() |
| **YOLOv4-Leaky** | 608 | 42.9 | 65.3 | 46.8 | 26.1 | 46.3 | 52.3 | [cfg](https://drive.google.com/open?id=1C4w_2loEpi-MznqMgKo16oZD7BcvgsCF) | [weights](https://drive.google.com/open?id=1dW-Sd70aTmXuFvYspiY85SwWdn0cr447) | [score]() |
| **YOLOv4-SAM-Leaky** | 608 | 43.3 | 65.4 | 47.1 | 26.2 | 46.1 | 53.2 | [cfg](https://drive.google.com/open?id=1gLCJI1KGD6mVdqy4K8jlxKHvwk1Texga) | [weights](https://drive.google.com/open?id=1HMJa5OUI4QG33ZkUEcyC8bfWDJUBB4qF) | [score]() |
| **YOLOv4-Mish** | 608 | 43.8 | 65.6 | 47.8 | 26.7 | 46.8 | 53.6 | [cfg](https://drive.google.com/open?id=1theQTAhnmtFQHtrnm7IFHbGHI6KGgKim) | [weights](https://drive.google.com/open?id=1sWNozS0emz7bmQTUWDLvsubLGnCwUiIS) | [score]() |
| **YOLOv4-SAM-Mish** | 608 | 44.0 | 65.7 | 48.0 | 26.7 | 47.2 | 54.1 | [cfg](https://drive.google.com/open?id=1V726wzWW1iBDcJ7uP8c0n_Y4M98lIg-C) | [weights](https://drive.google.com/open?id=1wK66ga9YgtjGNSm9fpouJn2GaDxa7SfT) | [score]() |
| **YOLOv4-CSP** | 608 | 44.2 | 65.8 | **48.2** | 26.7 | 47.2 | **54.7** | [cfg](https://drive.google.com/file/d/1XwtHTwn5mkS-Lqdn80PcULNyK-gmpEEm/view?usp=sharing) | [weights](https://drive.google.com/file/d/1FCGAQsBADMPdyfmZrOsSDeneikGWcnlu/view?usp=sharing) | [score]() |
| **YOLOv4-CSP-SAM** | 608 | **44.3** | **65.9** | **48.2** | **26.9** | **47.4** | 54.6 | [cfg](https://drive.google.com/file/d/1h7amsxwtWz6v66kzx53R3_Sxst5pti06/view?usp=sharing) | [weights](https://drive.google.com/file/d/1xBdS-r-iH44ZdmoXp5z5BlmHezzxNNUC/view?usp=sharing) | [score]() |

## COCO pre-trained files (416x416 random)

Models used anchors of 416x416 input resolution for training.

| Model | Size | AP | AP₅₀ | AP₇₅ | AP_S | AP_M | AP_L | cfg | weights | score |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| **YOLOv4-Leaky-416** | 416 | 40.7 | 62.7 | 43.9 | 21.4 | 43.7 | 54.0 | [cfg](https://drive.google.com/open?id=1vFz2f30ey6JyN6QbzRH7opS93AKXjOKf) | [weights](https://drive.google.com/open?id=1bV4RyU_-PNB78G-OtoTmw1Q7t_q90GKY) | [score]() |
| **YOLOv4-Mish-416** | 416 | **41.5** | **63.3** | **44.7** | **21.9** | **44.4** | **55.3** | [cfg](https://drive.google.com/open?id=1Ahmn7iI2B79jDlcKlRlD_tO5Tbr0Dq8x) | [weights](https://drive.google.com/open?id=1NuYL-MBKU0ko0dwsvnCx6vUr7970XSAR) | [score]() |

## COCO pre-trained files (416x416 resize)

Models used anchors of 416x416 input resolution for training.

| Model | Size | AP | AP₅₀ | AP₇₅ | AP_S | AP_M | AP_L | cfg | weights | score |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| **YOLOv4-tiny** | 416 | **20.2** | **40.1** | **8.4** | **7.1** | **23.4** | **27.4** | [cfg](https://raw.githubusercontent.com/AlexeyAB/darknet/master/cfg/yolov4-tiny.cfg) | [weights](https://github.com/AlexeyAB/darknet/releases/download/darknet_yolo_v4_pre/yolov4-tiny.weights) | [score]() |

## Select which model is suitable for training on your GPU

Here we list the minimum requirement GPU-RAM for training different models. However, we do not suggest you train the model with subdivisions equal or larger than 32, it will takes very long training time. We are developing the training method that can use less GPU-RAM to train a model.

* for 4GB-RAM GPUs
* YOLOv4-Leaky-416: `batch=64`, `subdivisions=64`.
* for 6GB-RAM GPUs
* YOLOv4-Leaky-416: `batch=64`, `subdivisions=32`.
* YOLOv4-Mish-416: `batch=64`, `subdivisions=64`.
* for 8GB-RAM GPUs
* YOLOv4-Mish-416: `batch=64`, `subdivisions=32`.
* for **11GB-RAM GPUs** (1080ti, 2080ti)
* **YOLOv4-Leaky-416**: `batch=64`, `subdivisions=16`.
* **YOLOv4-Mish-416**: `batch=64`, `subdivisions=16`.
* for **12GB-RAM GPUs** (titan x, titan xp, titan v)
* **YOLOv4-Leaky**: `batch=64`, `subdivisions=16`.
* **YOLOv4-SAM-Leaky**: `batch=64`, `subdivisions=16`.
* for **16GB-RAM GPUs** (p100)
* **YOLOv4**: `batch=64`, `subdivisions=16`.
* **YOLOv4-Mish**: `batch=64`, `subdivisions=16`.
* **YOLOv4-SAM-Mish**: `batch=64`, `subdivisions=16`.
* **YOLOv4-CSP**: `batch=64`, `subdivisions=16`.
* **YOLOv4-CSP-SAM**: `batch=64`, `subdivisions=16`.
* for **24GB-RAM GPUs** (titan rtx, rtx 6000)
* **YOLOv4-Leaky**: `batch=64`, `subdivisions=8`.
* **YOLOv4-SAM-Leaky**: `batch=64`, `subdivisions=8`.
* for **32GB-RAM GPUs** (v100, v100s)
* **YOLOv4**: `batch=64`, `subdivisions=8`.
* **YOLOv4-Mish**: `batch=64`, `subdivisions=8`.
* **YOLOv4-SAM-Mish**: `batch=64`, `subdivisions=8`.
* **YOLOv4-CSP**: `batch=64`, `subdivisions=8`.
* **YOLOv4-CSP-SAM**: `batch=64`, `subdivisions=8`.

## Model descriptions

* **YOLOv4**
* The model in [YOLOv4 paper]().
* **Backbone** - CSPDarknet53 with Mish activation
* **Neck** - PANet with Leaky activation
* **Plugin Modules** - SPP
* **V100 FPS** - 62@608x608, 83@512x512
* **BFLOPs** - 128.5@608x608
* **YOLOv4-Leaky**
* **Backbone** - CSPDarknet53 with Leaky activation
* **Neck** - PANet with Leaky activation
* **Plugin Modules** - SPP
* **BFLOPs** - 128.5@608x608
* **YOLOv4-SAM-Leaky**
* **Backbone** - CSPDarknet53 with Leaky activation
* **Neck** - PANet with Leaky activation
* **Plugin Modules** - SPP, SAM
* **BFLOPs** - 130.7@608x608
* **YOLOv4-Mish**
* **Backbone** - CSPDarknet53 with Mish activation
* **Neck** - PANet with Mish activation
* **Plugin Modules** - SPP
* **BFLOPs** - 128.5@608x608
* **YOLOv4-SAM-Mish**
* **Backbone** - CSPDarknet53 with Mish activation
* **Neck** - PANet with Mish activation
* **Plugin Modules** - SPP, SAM
* **V100 FPS** - 61@608x608, 81@512x512
* **BFLOPs** - 130.7@608x608
* **YOLOv4-CSP**
* **Backbone** - CSPDarknet53 with Mish activation
* **Neck** - CSPPANet with Mish activation
* **Plugin Modules** - CSPSPP
* **BFLOPs** - 112.4@608x608
* **YOLOv4-CSP-SAM**
* **Backbone** - CSPDarknet53 with Mish activation
* **Neck** - CSPPANet with Mish activation
* **Plugin Modules** - CSPSPP, SAM
* **V100 FPS** - 63@608x608, 84@512x512
* **BFLOPs** - 114.6@608x608
* **YOLOv4-Tiny**
* **Backbone** - CSPVoVNet with Leaky activation
* **Neck** - FPN with Leaky activation
* **BFLOPs** - 6.9@416x416