Is YOLOv9 better than YOLOv5 (v5u)?

On published COCO accuracy, YOLOv9. It leads 5 of the 5 size tiers where both families publish a figure, by an average of 2.7 mAP. The gap is widest at the nano tier, where YOLOv9 is ahead by 4.0 mAP. YOLOv9 tops out at 55.6 mAP and YOLOv5 (v5u) at 53.2. "Better" still depends on your constraint, though: accuracy is one axis and parameter count, task coverage and licence are the others, all of which are in the table above.

What is the difference between YOLOv9 and YOLOv5 (v5u)?

YOLOv9 (Academia Sinica (Wang et al.), 2024) and YOLOv5 (v5u) (Ultralytics, 2020) are 4 years apart. Fixes deep-network information loss with PGI and GELAN. The classic that made YOLO easy to ship, retrained anchor-free. Task coverage differs too: YOLOv9 supports 2 (Detect, Segment) against YOLOv5 (v5u)'s 3 (Detect, Segment, Classify).

Which is smaller, YOLOv9 or YOLOv5 (v5u)?

YOLOv9 gets more out of each parameter. YOLOv9c reaches 53.0 mAP on 25.5M parameters, matching or beating YOLOv5lu at 53.2M: about 52% fewer parameters for the same accuracy. That is the comparison worth making, since it is the same result from a smaller checkpoint.

Is YOLOv9 faster than YOLOv5 (v5u)?

Nothing on this page is a speed measurement, and it would be dishonest to imply one: every figure here is published COCO accuracy and parameter count. Parameters are a rough guide to inference cost within one family, but they do not transfer across architectures, and the post-processing step matters as much as the backbone. Export both to ONNX and measure on your own hardware and footage, which is the only comparison that predicts what you will actually get.

Are YOLOv9 and YOLOv5 (v5u) free for commercial use?

They differ: YOLOv9 is GPL-3.0 and YOLOv5 (v5u) is AGPL-3.0. That is a real decision, not a footnote, because the obligations are not the same and one of them may rule a model out of a closed-source product entirely. Read the licence text before you ship; this is a plain reading of it, not legal advice.

Model comparison · YOLOv9 vs YOLOv5 (v5u)

YOLOv9 vs YOLOv5 (v5u)

A size-tier-by-size-tier comparison of YOLOv9 and Ultralytics YOLOv5 (v5u): published COCO accuracy, parameter counts, licensing and which one to reach for. Every figure is the model authors' own, linked to its source.

CompareYOLOv9

AgainstYOLOv5 (v5u)

52%fewer params

mAP comparison

YOLOv9YOLOv5 (v5u)

YOLOv9

Nano 38.3
Small 46.8
Medium 51.4
Large 53.0
Extra large 55.6

YOLOv5 (v5u)

Nano 34.3
Small 43.0
Medium 49.0
Large 52.2
Extra large 53.2

COCO mAP at 640px, as published by each model’s authors. The y axis spans 31.0 to 59.0 rather than starting at zero, so small differences stay visible. Points are the published checkpoints; the joining lines are for reading order only.

Accuracy and size, tier by tier

Each row pairs the equivalent rung of both size ladders, so the numbers sit side by side at comparable capacity. Every figure is the model authors’ own.

COCO mAP50-95 @ 640px
mAP: higher is better
Params (M): lower is better

YOLOv9 and YOLOv5 (v5u) compared at each size tier: published COCO mAP50-95 at 640px and parameter counts in millions.
Tier	YOLOv9			YOLOv5 (v5u)			Δ mAP
Tier	Model	mAP	Params	Model	mAP	Params	Δ mAP
Nano	YOLOv9t	38.3	2.0M	YOLOv5nu	34.3	2.6M	+4.0
Small	YOLOv9s	46.8	7.2M	YOLOv5su	43.0	9.1M	+3.8
Medium	YOLOv9m	51.4	20.1M	YOLOv5mu	49.0	25.1M	+2.4
Large	YOLOv9c	53.0	25.5M	YOLOv5lu	52.2	53.2M	+0.8
Extra large	YOLOv9e	55.6	58.1M	YOLOv5xu	53.2	97.2M	+2.4

Δ mAP is YOLOv9 minus YOLOv5 (v5u): positive means YOLOv9 is ahead at that tier. Sources: YOLOv9 docs · YOLOv9 paper · YOLOv5 (v5u) docs

Additional information

YOLOv9 posts the higher published COCO mAP at all 5 matched size tiers, averaging 2.68 mAP ahead of YOLOv5 (v5u). The gap is widest at the nano tier, where YOLOv9 leads by 4.0 mAP. At the top of each range, YOLOv9 reaches 55.6 mAP against 53.2.
52% fewer parameters, same accuracy. YOLOv9c reaches 53.0 mAP from 25.5M parameters, matching or beating YOLOv5lu at 52.2 mAP from 53.2M.
Parameter count is not latency. It is a good proxy for download size and memory, but throughput depends on your hardware, batch size and export format, so measure on your own target.

YOLOv9

Fixes deep-network information loss with PGI and GELAN.

Released: 2024 · Academia Sinica (Wang et al.)
License: GPL-3.0
Framework: Original repo + Ultralytics (PyTorch)
Tasks: Detect, Segment

Adds Programmable Gradient Information and the GELAN backbone to preserve detail through deep layers, reaching high accuracy with very few parameters.

Download YOLOv9 weights →

YOLOv5 (v5u)

The classic that made YOLO easy to ship, retrained anchor-free.

Released: 2020 · Ultralytics
License: AGPL-3.0
Framework: Ultralytics (PyTorch)
Tasks: Detect, Segment, Classify

The anchor-free v5u retrain of the classic PyTorch YOLO, keeping its easy training and export while adopting the newer YOLOv8 detection head for better accuracy.

Download YOLOv5 (v5u) weights →

YOLOv9 and YOLOv5 (v5u) against the whole lineage

YOLOv9 and YOLOv5 (v5u) against every other detector here, on one shared scale. The rest are dimmed; hover any row to bring it back.

Every YOLO detector in this section, with release year, publisher, published COCO mAP50-95 range at 640px, parameter-count range and licence.
Model	Released	mAP range	Params range	Licence
YOLO26	2026	40.9 to 57.5	2.4M to 55.7M	AGPL-3.0
YOLO12	2025	40.6 to 55.2	2.6M to 59.1M	AGPL-3.0
YOLO11	2024	39.5 to 54.7	2.6M to 56.9M	AGPL-3.0
YOLOv10	2024	38.5 to 54.4	2.3M to 29.5M	AGPL-3.0
YOLOv9	2024	38.3 to 55.6	2.0M to 58.1M	GPL-3.0
YOLOv8	2023	37.3 to 53.9	3.2M to 68.2M	AGPL-3.0
YOLOv5 (v5u)	2020	34.3 to 53.2	2.6M to 97.2M	AGPL-3.0

Licenses here cover the weights. Running either model through the Ultralytics package brings AGPL-3.0 with it, and that can reach your own application regardless of how the checkpoint itself is licensed. Check the terms against your distribution plan before you ship.

Frequently asked questions

Is YOLOv9 better than YOLOv5 (v5u)?: On published COCO accuracy, YOLOv9. It leads 5 of the 5 size tiers where both families publish a figure, by an average of 2.7 mAP. The gap is widest at the nano tier, where YOLOv9 is ahead by 4.0 mAP. YOLOv9 tops out at 55.6 mAP and YOLOv5 (v5u) at 53.2. "Better" still depends on your constraint, though: accuracy is one axis and parameter count, task coverage and licence are the others, all of which are in the table above.
What is the difference between YOLOv9 and YOLOv5 (v5u)?: YOLOv9 (Academia Sinica (Wang et al.), 2024) and YOLOv5 (v5u) (Ultralytics, 2020) are 4 years apart. Fixes deep-network information loss with PGI and GELAN. The classic that made YOLO easy to ship, retrained anchor-free. Task coverage differs too: YOLOv9 supports 2 (Detect, Segment) against YOLOv5 (v5u)'s 3 (Detect, Segment, Classify).
Which is smaller, YOLOv9 or YOLOv5 (v5u)?: YOLOv9 gets more out of each parameter. YOLOv9c reaches 53.0 mAP on 25.5M parameters, matching or beating YOLOv5lu at 53.2M: about 52% fewer parameters for the same accuracy. That is the comparison worth making, since it is the same result from a smaller checkpoint.
Is YOLOv9 faster than YOLOv5 (v5u)?: Nothing on this page is a speed measurement, and it would be dishonest to imply one: every figure here is published COCO accuracy and parameter count. Parameters are a rough guide to inference cost within one family, but they do not transfer across architectures, and the post-processing step matters as much as the backbone. Export both to ONNX and measure on your own hardware and footage, which is the only comparison that predicts what you will actually get.
Are YOLOv9 and YOLOv5 (v5u) free for commercial use?: They differ: YOLOv9 is GPL-3.0 and YOLOv5 (v5u) is AGPL-3.0. That is a real decision, not a footnote, because the obligations are not the same and one of them may rule a model out of a closed-source product entirely. Read the licence text before you ship; this is a plain reading of it, not legal advice.

Every matchup

Looking for the weights themselves? Download any of these checkpoints from the model-weights hub, or see how the trackers behind them compare in the YOLO tracker benchmarks.