Skip to content

[HW Accel Support]: Docker Compose and Installation with NVIDIA 5070 #18169

Discussion options

You must be logged in to vote

Frigate 0.16 uses a slightly newer trt version, but in general if that doesn't work then you are talking about a major version upgrade to tensorrt 10 which is going to be a lot of refactoring.

Realistically though, what you should be able to do instead is just use 0.16 and then you can generate the onnx model (instead of converting to tensorrt) and run via onnx which should work fine on the 5070. If you do decide to go this way keep in mind that 0.16 is still in beta. There are dev specific docs which walk through how to download these models:

https://deploy-preview-16390--frigate-docs.netlify.app/configuration/object_detectors#onnx

Replies: 2 comments 9 replies

Comment options

You must be logged in to vote
8 replies
@angeres1
Comment options

@hawkeye217
Comment options

Answer selected by NickM-27
@angeres1
Comment options

@NickM-27
Comment options

NickM-27 May 12, 2025
Collaborator Sponsor

@hawkeye217
Comment options

@angeres1
Comment options

Comment options

You must be logged in to vote
1 reply
@angeres1
Comment options

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment