A Reflection of 2024s YOLOs!
Hello guys, hope you are having an awesome week.
A Reflection of 2024s YOLOs!
Hello guys, hope you are having an awesome week.
As we are waning down on 2024, a LinkedIn post by Muhammad Rizwan Munawar, Computer Vision Engineer at Ultralytics has indicated that YOLOv11 will be the last YOLO model class that will end of 2024 into 2025.
So let’s take a moment and look back at the models done in YOLO-verse for 2024!
February 2024- YOLOv9
This was the big model released first in 2024. Developed by people in Taiwan, YOLOv9 was a big upgrade from YOLOv8 by Ultralytics. They introduce new formats in image models: Programmable Gradient Information (PGI) and the Generalized Efficient Layer Aggregation Network (GELAN).
This model definitely had buffs in both speed of inference and inference accuracy.
And what is great is that there is even an image segmentation model provided by them!
YOLOv9 saw improvements overall in large margins in object detection and image segmentation!
Really YOLOv9 set the stage for YOLO models for the rest of the year!
May 2024- YOLOv10
Right on the heels of YOLOv9, researchers at Tsinghua University developed YOLOv10 directly on the Ultralytics library.
YOLOv10 saw many buffs in latency but not as much in accuracy, with the removal of NMS layers in the framework.
Other than this, YOLOv10, did not feature many accuracy buffs. However this was a big step forward in speeding up models and set the stage for YOLOv11.
Another big flaw with YOLOv10 was that it only provided image object detection and not segmentation. However, it was kinda justified with the amount of models produced which many people greatly appreciated.
And finally….
October 2024- YOLOv11
This release did not come as a surprise and ignored. Many people speculated a new model to be released when YOLO Vision Event by Ultralytics was announced. However, many people, me included, did not expect so many models to be refreshed.

YOLOv11 image object detection alone did not have that much improvements and was very small. However, YOLOv11 brought a refresh between image segmentation, pose, OBB, and classification! These statistics also came with large improvements compared to YOLOv8.
And additionally, Ultralytics released enterprise support and models. I don’t have that much information about it as an individual programmer, but I assume these come with a lot of Ultralytics support.
Yeah and that is about it. There were many other models that were released, however they were small scale and not one of the few major versions developed in the YOLO-verse. 2024 brought a lot of good models to the YOLO-verse and I have high hopes to 2025.
And that’s about it. If you are here, you probably love my articles. So go ahead and clap my article if you liked it, give your comments about 2024 YOLO models, and check out my other articles which you will probably enjoy! Additionally subscribe to my Medium to be updated on the new YOLO news and other computer science, techy, and political stuff here. Follow as well.
And that is about it, I will catch you on the flip-flop. I’m out, see yah!
메타데이터
- post_id
- d94a27fdbc98
- slug
- a-reflection-of-2024s-yolos-d94a27fdbc98
- url
- https://medium.com/@zainshariff6506/a-reflection-of-2024s-yolos-d94a27fdbc98
- canonical_url
- https://medium.com/@zainshariff6506/a-reflection-of-2024s-yolos-d94a27fdbc98
- author_url
- https://medium.com/@zainshariff6506
- status
- ok
- fetched_at
- 2026-07-21 17:51:44