YOLOv8-RKNN2-Ultralytics

Looking for the Rockchip Model Zoo and custom exports? See YOLOv8-RKNN2.

100 FP16 and INT8 YOLOv8 exports for RK3576 and RK3588, produced with the Ultralytics export workflow and task-specific RKNN conversion tools. Supported tasks include classification, detection, oriented bounding boxes, pose, and instance segmentation.

Choose a model

Models are organized by toolkit version, task, variant, target chip, and precision:

models/v<toolkit-version>/<task>/<variant>/<chip>/<precision>/

Download model.rknn and metadata.yaml from the same directory. The manifest lists the available models and checksums.

INT8 detection

Use the metadata.yaml from the same model directory. INT8 detection outputs normalized box coordinates and requires coordinates: normalized. Import both files together in the warehouse, or keep them in the same directory for the CLI. Treating normalized coordinates as pixels can result in no boxes.

Reproduce the conversion

The source directory contains the original checkpoints, conversion scripts, and environment and calibration records. Conversion used RKNN Toolkit2 2.3.2; upstream Ultralytics source is available on PyPI.

python source/convert-model.py --help

For INT8 detection, pass a COCO128 dataset YAML. For other INT8 tasks, pass a COCO128 image list. Use --calibration-limit 128 and the input size listed in the manifest.

Source and license

The inputs are official Ultralytics checkpoints. Conversion changes the format and quantizes the INT8 graphs. The RKNN files and original weights are distributed under AGPL-3.0; see LICENSE and the checkpoint URLs and hashes.