FocsleFocsle
BR

RT-DETR-Edge

22%
by Baidu Research

Real-time DETR optimized for transformer-friendly NPUs. Wins on transformer-class silicon, struggles on legacy NPUs.

Object detectionApache-2.0INT8FP16transformercocodetection
92K downloads 5.4K deploymentsUpdated Apr 2, 2028
Headline:18.6ms · NVIDIA Jetson Orin Nano · FP16

Files

Weights, configs, and per-vendor compiled exports. Files marked LFS are stored on the Fo’c’sle CDN and served regional.

NameSizeUpdated
README.md8.4 KB2028-04-02
config.json2.1 KB2028-04-02
tokenizer.json1.8 MB2028-04-02
model.safetensorsLFS25.7 MB2028-04-02
rt-detr-edge.onnxLFS22.9 MB2028-04-02
rt-detr-edge_int8.tfliteLFS22 MB2028-04-02
rt-detr-edge.jetson-orin-nano.planLFS35 MB2028-04-02
rt-detr-edge.jetson-agx-orin.planLFS35 MB2028-04-02
rt-detr-edge.jetson-thor.planLFS35 MB2028-04-02
rt-detr-edge.mlmodelcLFS24 MB2028-04-02
examples/2028-04-02
benchmarks/2028-04-02
LICENSE11.4 KB2028-04-02