YOLO26-RKNN2
Looking for the Ultralytics RKNN exports? See YOLO26-RKNN2-Ultralytics.
YOLO26 RKNN exports for RK3576 and RK3588, compiled with RKNN Toolkit2 2.3.2. Available tasks: detection, classification, depth, oriented bounding boxes, pose, instance segmentation, and semantic 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.
Conversion
Detection models were converted on Seeed-Dev with RKNN Toolkit2 2.3.2 using ONNX inputs supplied by the Rockchip Model Zoo. The manifest records input ONNX and output RKNN checksums. The ONNX graphs and Model Zoo download URLs are in source/onnx; the graphs declare their Ultralytics exporter versions and AGPL-3.0 license.
Source and license
Original checkpoints, conversion scripts, and provenance are in source. The models and checkpoints are distributed under AGPL-3.0; see LICENSE and checkpoint details.