administrator/frigate
1
0
Fork 0
mirror of https://github.com/blakeblackshear/frigate.git synced 2026-03-17 21:58:22 +03:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

Update inference times for RTX 5060 Ti (#22423)

Browse Source 
* Update inference times for RTX 5060 Ti

* Ad RF-DETR

Added Nano-320 performance data for RTX 5060 Ti.
This commit is contained in:
Nicolas Mowen 2026-03-13 10:43:16 -06:00 committed by GitHub
parent 324953d3a5
commit 7864c446fc
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 9 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
docs/docs/frigate/hardware.md
View File

@ -200,13 +200,14 @@ Inference is done with the `onnx` detector type. Speeds will vary greatly depend
✅ - Accelerated with CUDA Graphs
❌ - Not accelerated with CUDA Graphs
| Name | ✅ YOLOv9 Inference Time | ✅ RF-DETR Inference Time | ❌ YOLO-NAS Inference Time |
| --------- | ------------------------------------- | ------------------------- | -------------------------- |
| GTX 1070 | s-320: 16 ms | | 320: 14 ms |
| RTX 3050 | t-320: 8 ms s-320: 10 ms s-640: 28 ms | Nano-320: ~ 12 ms | 320: ~ 10 ms 640: ~ 16 ms |
| RTX 3070 | t-320: 6 ms s-320: 8 ms s-640: 25 ms | Nano-320: ~ 9 ms | 320: ~ 8 ms 640: ~ 14 ms |
| RTX A4000 | | | 320: ~ 15 ms |
| Tesla P40 | | | 320: ~ 105 ms |
| Name | ✅ YOLOv9 Inference Time | ✅ RF-DETR Inference Time | ❌ YOLO-NAS Inference Time |
| ----------- | ------------------------------------- | ------------------------- | -------------------------- |
| GTX 1070 | s-320: 16 ms | | 320: 14 ms |
| RTX 3050 | t-320: 8 ms s-320: 10 ms s-640: 28 ms | Nano-320: ~ 12 ms | 320: ~ 10 ms 640: ~ 16 ms |
| RTX 3070 | t-320: 6 ms s-320: 8 ms s-640: 25 ms | Nano-320: ~ 9 ms | 320: ~ 8 ms 640: ~ 14 ms |
| RTX 5060 Ti | t-320: 5 ms s-320: 7 ms s-640: 22 ms | Nano-320: ~ 6 ms | |
| RTX A4000 | | | 320: ~ 15 ms |
| Tesla P40 | | | 320: ~ 105 ms |
### Apple Silicon
@ -320,4 +321,4 @@ Basically - When you increase the resolution and/or the frame rate of the stream
YES! The Coral does not help with decoding video streams.
Decompressing video streams takes a significant amount of CPU power. Video compression uses key frames (also known as I-frames) to send a full frame in the video stream. The following frames only include the difference from the key frame, and the CPU has to compile each frame by merging the differences with the key frame. [More detailed explanation](https://support.video.ibm.com/hc/en-us/articles/18106203580316-Keyframes-InterFrame-Video-Compression). Higher resolutions and frame rates mean more processing power is needed to decode the video stream, so try and set them on the camera to avoid unnecessary decoding work.
Decompressing video streams takes a significant amount of CPU power. Video compression uses key frames (also known as I-frames) to send a full frame in the video stream. The following frames only include the difference from the key frame, and the CPU has to compile each frame by merging the differences with the key frame. [More detailed explanation](https://support.video.ibm.com/hc/en-us/articles/18106203580316-Keyframes-InterFrame-Video-Compression). Higher resolutions and frame rates mean more processing power is needed to decode the video stream, so try and set them on the camera to avoid unnecessary decoding work.