DetectFPS
Submit a build

Rockchip RK3588 NPU

AcceleratorRKNN (Rockchip)

6 builds on DetectFPS: 5 Frigate docs references and 1 community build. All 6 are usable for stats. Hardware id rockchip-rk3588.

Estimate capacityCompare Submit a build with this device
Median inference
25 ms
all models · 6 usable builds
Fastest model
14 ms
yolox-nano · 1 docs reference
Most cameras seen
1
db-128 · yolov8
Price
not tracked
no verified price yet
Max detectors
—
no limit recorded

Median inference by model

Capacity and cameras are estimates
Model · inputMedianCapacityCameras
yolox-nano
14 ms
1 docs reference
71 det/s
17
yolox-tiny
18 ms
1 docs reference
56 det/s
13
yolo-nas-s
20 ms
1 docs reference
50 det/s
12
yolo-nas-m
30 ms
1 docs reference
33 det/s
8
yolov9-t
35 ms
1 docs reference
29 det/s
6
yolov8
40 ms
1 community build
25 det/s
6

Bars show relative throughput (fastest model = full). Capacity = 1000 ÷ median ms for one detector. Cameras = how many at 5 fps and Typical activity stay under 60% of that capacity. Where a row mixes docs and community numbers, treat the median as a range rather than a point.

All builds for this device

Open in the database →
BuildSourceModelSpeedCamsFrigatePosted
db-128 community yolov8 40 ms 1 0.12.1 Sep 2023 · GitHub #7601
db-103 Frigate docs yolox-nano 14 ms — — checked 2026-10-19
db-104 Frigate docs yolox-tiny 18 ms — — checked 2026-10-19
db-101 Frigate docs yolo-nas-s 20 ms — — checked 2026-10-19
db-102 Frigate docs yolo-nas-m 30 ms — — checked 2026-10-19
db-100 Frigate docs yolov9-tiny 35 ms — — checked 2026-10-19

Community builds first, then docs references.