DX-Model Zoo

Pre-quantized, pre-compiled networks benchmarked on the DX-M1 NPU. Every entry ships an ONNX source, a compiled .dxnn binary, and measured throughput.

Benchmark
environment

Results may vary with the specific hardware configuration.

NPU
DX-M1 M.2 module
Host CPU
Intel(R) Core(TM) i5-14600K, 32G RAM
SDK version
dx-com v2.4.0, dx-rt v3.4.0
Power measurement
DX-M1 NPU chip only
Benchmark cmd
dxrun -m <MODEL_FILE> --use-ort -t 5 -b

Coverage by models
Quantization tier
{{ modeCount }} of {{ total }} models publish a {{ modeLabel }} result
Get model
{{ r.display }} {{ r.task }} {{ r.input }} {{ r.params }} {{ r.acc }} {{ r.delta }} {{ r.fps }} {{ r.fpsw }} {{ r.license }}

No models match these filters.

{{ chartTitle }}

{{ chartCaption }}{{ chartCaptionGap }}{{ zoomHint }}

{{ chart }}

License Notice

Please review the following before downloading the model:

  • I have reviewed the license terms for the model I want to download.
  • If I distribute a commercial product using this model, I may be obligated to disclose my source code.
  • DEEPX is not responsible for any license disputes arising from the use of this model.
{{ pendingName }}