Roboflow

YOLOv4 vs YOLOv9

We don't benchmark YOLOv4 or YOLOv9 on Vision Evals. YOLOv4 is open-weight under Custom. YOLOv9 is open-weight under GPL v3: commercial use means releasing your source code under GPL-3.0.

Compare YOLOv4 vs YOLOv9 live

Run the same image across every model that supports a task and compare their outputs side-by-side.

These models don't share enough common tasks for a side-by-side demo. See the comparison table below for their capabilities.

YOLOv4 vs YOLOv9 Comparison Table

PropertyYOLOv4YOLOv9
OrganizationAcademia SinicaAcademia Sinica
Categoryopenopen
Modalityvisionvision
Release DateApr 2020Feb 2024
Context Window——
ParametersUnknown2.0M-57.3M
LicenseCustomGPL v3
Commercial useSee license termscommercial use means releasing your source code under GPL-3.0
Vision Tasks
Instance SegmentationNot listedSupported
Both support 1 more task
Model Features
Real-Time VisionNot listedSupported

YOLOv4 vs YOLOv9: Overview

YOLOv4

YOLOv4 is an object detection model developed by Alexey Bochkovskiy, Chien-Yao Wang, and Hong-Yuan Mark Liao at Academia Sinica, released in April 2020 via the Darknet framework. It combines a CSPDarknet53 backbone, PANet neck, and YOLOv3 detection head with a large set of training improvements — Bag of Freebies and Bag of Specials — that improve accuracy with minimal inference cost increase.

YOLOv4 achieves 43.5% AP on COCO at 65 FPS on a Tesla V100 GPU. The Darknet implementation is the original version, distinguishing it from subsequent PyTorch-based reimplementations. It remains a widely referenced detection architecture and a supported training target in Roboflow Inference.

YOLOv9

YOLOv9 is a real-time object detection model developed by Chien-Yao Wang and Hong-Yuan Mark Liao at Academia Sinica, released in February 2024 under the GPL-3.0 license. It introduces Programmable Gradient Information (PGI), a mechanism that preserves complete input information through auxiliary reversible branches during training to address information loss in deep network layers. It also introduces the Generalized Efficient Layer Aggregation Network (GELAN), which achieves better parameter utilization compared to prior CSP-based designs.

YOLOv9-C achieves 53.0% AP on COCO with 42% fewer parameters and 21% less computation than YOLOv8-C at comparable accuracy. YOLOv9-E achieves 55.6% AP. The model is deployable through Roboflow Inference and supports fine-tuning via the standard training pipeline in the official repository.