Aug 25, 2026

FastestDetV2: a Light-weight Object Detector

🔥🔥Even faster and stronger than FastestDet | 比FastestDet更快更强🔥🔥 - Pairman/FastestDetV2

github.com

GitHub - Pairman/FastestDetV2: 🔥🔥Even faster and stronger than FastestDet | 比FastestDet更快更强🔥🔥

Improvements

2.5% mAP50 & 1% mAP50:95 improvement, with ~20% faster speed compared to FastestDet
Assign Guidance Module and SimOTA label assignment for better precision
Quantization-aware, reparameterizable MobileOne backbone and convolution modules

Gallery

Benchmarks

Model	mAP50	mAP50:95	Resolution	Inference time (4x core)	Inference time (1x core)	Params (M)
FastestDetV2	27.8%	14.0%	352X352	2.83ms	6.95ms	0.33M
FastestDetV2-2x	36.6%	19.9%	352X352	6.81ms	19.88ms	1.22M
FastestDet	25.3%	13.0%	352X352	3.68ms	8.48ms	0.24M
NanoDet-m	-	20.6%	320X320	7.76ms	22.23ms	0.95M
YOLOX-Nano	-	25.8%	416X416	36.88ms	92.52ms	0.91M
YOLOv8n	56.8%	37.4%	640X640	57.03ms	122.63ms	7.2M

Tested on EmbedFire LubanCat-4 RK3588S ARM 4*Cortex-A76 CPU@2.0GHz, using NCNN.

Multi-platform Benchmarks

Device	Computing backend	System	Framework	Inference time (4x core)	Inference time (1x core)	2x Inference time (4x core)	2x Inference time (1x core)
Huawei Atlas 800I A3	Ascend 910_9362 (NPU)	Linux (arm64)	CANN	/	0.45ms	/	0.58ms
EmbedFire LubanCat-4	RK3588 (CPU) ¹	Linux (arm64)	NCNN	2.83ms	6.95ms	6.81ms	19.88ms
EmbedFire LubanCat-4	RK3588 (NPU)	Linux (arm64)	RKNN	7.067ms ²	7.532ms	8.04ms ³	9.56ms
Google Pixel 10 Pro XL	Tensor G5 (CPU)	Android (arm64)	NCNN	2.69ms	3.88ms	4.66ms	6.26ms
OnePlus 6	Snapdragon 845 (CPU)	Android (arm64)	NCNN	4.73ms	8.14ms	11.56ms	17.84ms
Dell Precision 3630 Tower	Core i9-9900 (CPU) ⁴	Linux (x86_64)	NCNN	2.90m	7.31ms	6.86ms	19.94ms

¹: At 2.0 GHz.
², ³: RKNNLite.NPU_CORE_0_1_2 is used.
⁴: At 800MHz.

Model Zoo

Download	Note
fastestdetv2.pth, fastestdetv2_unfused.pth fastestdetv2-2x.pth, fastestdetv2-2x_unfused.pth	Model weights
qamobileone.pth qamobileone-2x.pth	Backbone weights
fastestdetv2.apk	Android demo
fastestdetv2.bin, fastestdetv2.param fastestdetv2-2x.bin, fastestdetv2-2x.param	NCNN files
fastestdetv2.onnx fastestdetv2-2x.onnx	ONNX files
(target platform-specific, not provided)	CANN files
(target platform-specific, not provided)	RKNN files
fastestdetv2.pt, fastestdetv2_ptq.arm.pt, fastestdetv2_ptq.x86.pt fastestdetv2-2x.pt, fastestdetv2-2x_ptq.arm.pt, fastestdetv2-2x_ptq.x86.pt	TorchScript files

Usage

Dependencies

pip install -r requirements.txt

Datasets & Configurations

Datasets can be either in Darknet format (like FastestDet, using a text file to list image paths, with labels stored in separate .txt files in the same directory) or in YOLO format (like YOLOv8, where each image has a corresponding .txt label file in a seperate directory). Labels are in cls cx cy w h normalized bboxes.

The .yaml configurations file specifies dataset paths, model settings, and training hyperparameters. Dataset could be either in Darknet format or YOLO format. Class names can also be in a single text file with each line representing a class name. See configs/coco.yaml for example.

Evaluation & Testing

You can evaluate the model with a fused (reparameterized) model weights file.

python3 eval.py --configs CONFIGS_PATH --weight WEIGHTS_PATH

Or test it on an image:

python3 test.py --configs CONFIGS_PATH --weights WEIGHTS_PATH --image IMAGE_PATH

Training

Download the backbone weights and place it under weights/qamobileone.pth and weights/qamobileone-2x.pth, and run:

python3 train.py --configs CONFIGS_PATH

Or finetune it with an unfused weights file:

python3 train.py --configs CONFIGS_PATH --weights WEIGHTS_PATH

Deployment

ONNX & TorchScript

Export to ONNX and TorchScript format with:

python3 test.py --configs CONFIGS_PATH --weights WEIGHTS_PATH --export

PT2E PTQ

Post-training quantization for x86 (with X86InductorQuantizer) or arm (with XNNPackQuantizer) platforms, with fused weights:

python3 quant.py --configs CONFIGS_PATH --weights WEIGHTS_PATH --image IMAGE_PATH --target TARGET_PLATFORM

Citation

@misc{=FastestDetV2,
    title={FastestDetV2: Even faster and stronger than FastestDet},
    author={Pairman},
    howpublished = {\url{https://github.com/Pairman/FastestDetV2}},
    year={2025}
}

References

FastestDet: https://github.com/dog-qiuqiu/FastestDet
Assign Guidance Module and NCNN deployment: https://github.com/RangiLyu/nanodet
MobileOne: https://github.com/apple/ml-mobileone and https://github.com/glory-wan/TF-Net
Quantization-aware RepConv: https://github.com/meituan/YOLOv6
SimOTA label assignment: https://github.com/open-mmlab/mmyolo
NCNN: https://github.com/Tencent/ncnn