Model weights · YOLO-World

Download YOLO-World weights

Fuses a YOLO detector with CLIP text embeddings, so you can detect any object by typing its name, with no retraining. This is called open-vocabulary detection. Pick a size below and download the official .pt checkpoint in one click, with published parameters, size and license all in view.

GPL-3.0Tencent AI Lab · 2024 · Ultralytics (PyTorch)

YOLO-World checkpoints, pick a size and download

Fuses a YOLO detector with CLIP text embeddings, so you can detect any object by typing its name, with no retraining. This is called open-vocabulary detection.

  • YOLOv8s-worldv2
    yolov8s-worldv2.pt
    Params
    12.7M
  • YOLOv8m-worldv2
    yolov8m-worldv2.pt
    Params
    28.4M
  • YOLOv8l-worldv2
    yolov8l-worldv2.pt
    Params
    46.8M
  • YOLOv8x-worldv2
    yolov8x-worldv2.pt
    Params
    72.9M

Figures are published by the authors. Weights host: github.com. Clicking Download verifies the file and starts it straight from the official CDN.

How to load YOLO-World weights

Install the Ultralytics package, then point the loader at the checkpoint. It downloads automatically on first use, or you can pass the local path to the file you downloaded above.

pip install ultralytics
from ultralytics import YOLO

# Downloads on first use, or pass a local path to your .pt file
model = YOLO("yolov8s-worldv2.pt")
results = model("image.jpg")

Want to see the architecture? Export to ONNX and open it in the ONNX visualizer for a labelled diagram with shapes and parameter counts.

Open visualizer →

Choosing a YOLO-World checkpoint

Read off the table above: what each step up the size ladder costs and returns for this family specifically, where it stops being worth it, and what the license actually permits.

What the size ladder buys

YOLO-World ships 4 checkpoints, from YOLOv8s-worldv2 at 12.7M parameters to YOLOv8x-worldv2 at 72.9M, 5.7× the model. These are promptable models with no single comparable accuracy figure, so the table lists parameters and download size rather than inventing one. Size and throughput are the axis you choose on: pick the smallest checkpoint that holds up on your own data rather than the largest you can fit.

Licensing, in practice

The weights are GPL-3.0. GPL-3.0 requires derivative works to be released under GPL too.

About YOLO-World

YOLO-World (Tencent AI Lab) brings open-vocabulary detection to the YOLO speed regime: give it class names as text and it detects them zero-shot, no retraining required. It fuses a YOLOv8 detector with CLIP text embeddings via a re-parameterizable vision-language path. The v2 checkpoints below are the recommended, more accurate release.

Author
Tencent AI Lab
Released
2024
Tasks
Open-vocab detect
Framework
Ultralytics (PyTorch)
Input size
640px
License
GPL-3.0
Resources
DocsGitHubPaperLicense

More model weights to download

Each page lists every checkpoint with accuracy, parameters, license and a one-click download.

See all 19 model families →

YOLO tracker FPS benchmarks →Get in touch →
Blog

From the blog

Tutorials, code, and notes on computer vision, deep learning, and applied AI.

View all posts →

Frequently asked questions

How do I download YOLO-World weights?
Click the Download button next to any variant in the table above; we verify the file and start it straight from the official CDN. You can also let the loader fetch it automatically on first use with `model = YOLO("yolov8s-worldv2.pt")`. All 4 YOLO-World checkpoints are hosted officially.
Is YOLO-World free for commercial use?
YOLO-World is released under GPL-3.0. GPL-3.0 requires derivative works to also be open-sourced under GPL; review it against your deployment. Always confirm against the linked license text.
Which YOLO-World model size should I use?
Start with YOLOv8s-worldv2, the smallest and fastest, ideal for prototyping, edge and CPU. Move up the ladder only when you need more accuracy and have the compute for it.