Is YOLO26 better than YOLO12?

On published COCO accuracy, YOLO26. It leads 5 of the 5 size tiers where both families publish a figure, by an average of 1.0 mAP. The gap is widest at the extra large tier, where YOLO26 is ahead by 2.3 mAP. YOLO26 tops out at 57.5 mAP and YOLO12 at 55.2. "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 YOLO26 and YOLO12?

YOLO26 (Ultralytics, 2026) and YOLO12 (Tian, Ye & Doermann (UB / UCAS), 2025) are 1 year apart. NMS-free, end-to-end, and tuned for the edge. The attention-centric generation, between YOLO11 and YOLO26. Task coverage differs too: YOLO26 supports 8 (Detect, Segment, Semantic segmentation, Depth, Classify, Pose, OBB, Open-vocabulary) against YOLO12's 5 (Detect, Segment, Classify, Pose, OBB).

Which is smaller, YOLO26 or YOLO12?

Compare them tier by tier rather than family to family, since both ship a full ladder from nano to extra large and the size you pick matters far more than the generation. The parameter columns in the table above give the figure at every tier.

Is YOLO26 faster than YOLO12?

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 YOLO26 and YOLO12 free for commercial use?

Both ship under AGPL-3.0, so the licence is not a tiebreaker here: choose on accuracy and size. AGPL-3.0 is free for open-source, research and internal use, but a closed-source product or paid service needs an Ultralytics Enterprise licence. Read the licence text before you ship; this is a plain reading of it, not legal advice.

Model comparison · YOLO26 vs YOLO12

YOLO26 vs YOLO12

A size-tier-by-size-tier comparison of Ultralytics YOLO26 and YOLO12: published COCO accuracy, parameter counts, licensing and which one to reach for. Every figure is the model authors' own, linked to its source.

CompareYOLO26

AgainstYOLO12

8%fewer params

mAP comparison

YOLO26YOLO12

YOLO26

Nano 40.9
Small 48.6
Medium 53.1
Large 55.0
Extra large 57.5

YOLO12

Nano 40.6
Small 48.0
Medium 52.5
Large 53.7
Extra large 55.2

COCO mAP at 640px, as published by each model’s authors. The y axis spans 38.0 to 60.5 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

YOLO26 and YOLO12 compared at each size tier: published COCO mAP50-95 at 640px and parameter counts in millions.
Tier	YOLO26			YOLO12			Δ mAP
Tier	Model	mAP	Params	Model	mAP	Params	Δ mAP
Nano	YOLO26n	40.9	2.4M	YOLO12n	40.6	2.6M	+0.3
Small	YOLO26s	48.6	9.5M	YOLO12s	48.0	9.3M	+0.6
Medium	YOLO26m	53.1	20.4M	YOLO12m	52.5	20.2M	+0.6
Large	YOLO26l	55.0	24.8M	YOLO12l	53.7	26.4M	+1.3
Extra large	YOLO26x	57.5	55.7M	YOLO12x	55.2	59.1M	+2.3

Δ mAP is YOLO26 minus YOLO12: positive means YOLO26 is ahead at that tier. Sources: YOLO26 docs · YOLO12 docs · YOLO12 paper

Additional information

YOLO26 posts the higher published COCO mAP at all 5 matched size tiers, averaging 1.02 mAP ahead of YOLO12. The gap is widest at the extra large tier, where YOLO26 leads by 2.3 mAP. At the top of each range, YOLO26 reaches 57.5 mAP against 55.2.
8% fewer parameters at the nano tier. YOLO26n reaches 40.9 mAP from 2.4M parameters against YOLO12n at 40.6 mAP from 2.6M, so the saving costs 0.3 mAP. No checkpoint from either family matches a larger one from the other outright.
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.

YOLO26

NMS-free, end-to-end, and tuned for the edge.

Released: 2026 · Ultralytics
License: AGPL-3.0
Framework: Ultralytics (PyTorch)
Tasks: Detect, Segment, Semantic segmentation, Depth, Classify, Pose, OBB, Open-vocabulary

A single-stage detector that outputs final boxes directly, with no separate NMS cleanup step, so it runs faster on CPUs and edge devices while keeping top accuracy.

Download YOLO26 weights →

YOLO12

The attention-centric generation, between YOLO11 and YOLO26.

Released: 2025 · Tian, Ye & Doermann (UB / UCAS)
License: AGPL-3.0
Framework: Ultralytics (PyTorch)
Tasks: Detect, Segment, Classify, Pose, OBB

An attention-centric YOLO: area attention and residual ELAN blocks replace some of the pure-convolution stack, buying about a point of mAP over YOLO11 at a similar size.

Download YOLO12 weights →

YOLO26 and YOLO12 against the whole lineage

YOLO26 and YOLO12 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

Frequently asked questions

Is YOLO26 better than YOLO12?: On published COCO accuracy, YOLO26. It leads 5 of the 5 size tiers where both families publish a figure, by an average of 1.0 mAP. The gap is widest at the extra large tier, where YOLO26 is ahead by 2.3 mAP. YOLO26 tops out at 57.5 mAP and YOLO12 at 55.2. "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 YOLO26 and YOLO12?: YOLO26 (Ultralytics, 2026) and YOLO12 (Tian, Ye & Doermann (UB / UCAS), 2025) are 1 year apart. NMS-free, end-to-end, and tuned for the edge. The attention-centric generation, between YOLO11 and YOLO26. Task coverage differs too: YOLO26 supports 8 (Detect, Segment, Semantic segmentation, Depth, Classify, Pose, OBB, Open-vocabulary) against YOLO12's 5 (Detect, Segment, Classify, Pose, OBB).
Which is smaller, YOLO26 or YOLO12?: Compare them tier by tier rather than family to family, since both ship a full ladder from nano to extra large and the size you pick matters far more than the generation. The parameter columns in the table above give the figure at every tier.
Is YOLO26 faster than YOLO12?: 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 YOLO26 and YOLO12 free for commercial use?: Both ship under AGPL-3.0, so the licence is not a tiebreaker here: choose on accuracy and size. AGPL-3.0 is free for open-source, research and internal use, but a closed-source product or paid service needs an Ultralytics Enterprise licence. 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.