| license: other | |
| license_name: upstream-terms-by-artifact | |
| license_link: https://huggingface.co/HanzoHuang/YOLO-World-RKNN2/blob/main/LICENSES.md | |
| pipeline_tag: object-detection | |
| tags: | |
| - rknn | |
| - rk3576 | |
| - rk3588 | |
| # YOLO-World-RKNN2 | |
| > **Looking for Ultralytics RKNN exports?** Browse the [Ultralytics exports collection](https://huggingface.co/collections/HanzoHuang/ultralytics-rknn-exports-6af1d088ba04ccf2dcbf81c2). | |
| YOLO-World builds from the Rockchip Model Zoo for **RK3576** and **RK3588**. | |
| ## Choose a model | |
| Models are grouped by toolkit version, task, variant, chip, and precision: | |
| ```text | |
| models/v<toolkit-version>/<task>/<variant>/<chip>/<precision>/ | |
| ``` | |
| YOLO-World requires its paired CLIP text encoder and Rockchip’s demo. | |
| ## Toolkit versions | |
| - [Toolkit 2.5.0 builds](tree/main/models/v2.5.0) use RKNN Toolkit2 2.5.0. Conversion receipts identify the source ONNX model and converter. | |
| - [Toolkit 2.3.2 builds](tree/main/models/v2.3.2) were tested on their target board. Each conversion receipt identifies its source graph and compiler. | |
| See [source and license notices](LICENSES.md) for provenance and applicable terms. | |