[HW Accel Support]: Jetson Nano with frigate 0.16

[andrewhack]

Describe the problem you are having

As Nano is not supported by JP5 and JP6 and Frigate 0.16 is dropping JP4, is there are custom solution to use 0.16 with Jetson Nano?

Version

0.16.0-rc1

Frigate config file

.

docker-compose file or Docker CLI command

services:
  frigate:
    privileged: true
    environment:
      - YOLO_MODELS=yolov7-320
      - USE_FP16=false
    container_name: frigate
    runtime: nvidia
    volumes:
      - /home/frigate/config:/config
      - /home/frigate/media:/media
    ports:
      - "5000:5000"
      - "8554:8554"
    image: ghcr.io/blakeblackshear/frigate:0.16.0-rc1-tensorrt-jp6

Relevant Frigate log output

Creating yolov7-320.cfg and yolov7-320.weights

Done.

Generating yolov7-320.trt. This may take a few minutes.

Traceback (most recent call last):
  File "/usr/local/src/tensorrt_demos/yolo/onnx_to_tensorrt.py", line 57, in <module>
    import tensorrt as trt
  File "/usr/local/lib/python3.11/dist-packages/tensorrt/__init__.py", line 68, in <module>
    from .tensorrt import *
ImportError: libnvcudla.so: cannot open shared object file: No such file or directory
Available tensorrt models:
ls: cannot access '*.trt': No such file or directory
s6-rc: warning: unable to start service trt-model-prepare: command exited 2

Relevant go2rtc log output

.

FFprobe output from your camera

.

Install method

Docker Compose

Object Detector

TensorRT

Network connection

Wired

Camera make and model

.

Screenshots of the Frigate UI's System metrics pages

.

Any other information that may be helpful

No response

[inkeep-ai-assistant[bot]]

Frigate 0.16 does not officially support Jetson Nano, as JetPack 4 (the version compatible with Nano) is not supported in 0.16 and newer builds focus on JetPack 5 and 6, which are not compatible with Nano hardware(1)(2).

There is no documented custom or official solution to use Frigate 0.16 with Jetson Nano. The TensorRT detector and Jetson-specific hardware acceleration are only supported in builds for JetPack 5 and 6, which do not run on Nano(3)(4)(2)(1). Attempts to use a JP6-based image (such as ghcr.io/blakeblackshear/frigate:0.16.0-rc1-tensorrt-jp6) on Jetson Nano will fail due to incompatible libraries and missing hardware support, as shown in your log with missing libnvcudla.so(1).

Check out these potentially useful sources:
Hardware Acceleration,
Supported Hardware,
Recommended hardware

Would you like a general answer based on my knowledge instead?

📚 Sources:

1. Support for Nvidia Jetson JetPack 6
2. Recommended hardware
3. Hardware Acceleration
4. Supported Hardware

---

Have another question? Just tag @inkeep.