Thanks for the video Craig, I really enjoy using the AI gateway, is there any plans to support using custom endpoints? I have my own models that I’m deploying on a serverless GPU platform (for now till workers AI supports ONNX 😉). would love to track those calls with the AI gateway. right now i’m storing the logs in D1 but that’s painful
@ only for off the shelf providers. like google vertex ai, open ai, or similar. if you’re hosting your own models on ec2 or other gpu provider it doesn’t work
![[#2024MAMA] ROSÉ (로제), Bruno Mars - APT. | Mnet 241122 방송](http://i.ytimg.com/vi/dgGqD28J6aQ/mqdefault.jpg)
Great run down!
Thanks for the video Craig, I really enjoy using the AI gateway, is there any plans to support using custom endpoints? I have my own models that I’m deploying on a serverless GPU platform (for now till workers AI supports ONNX 😉). would love to track those calls with the AI gateway. right now i’m storing the logs in D1 but that’s painful
its in Universal Endpoint
@ only for off the shelf providers. like google vertex ai, open ai, or similar. if you’re hosting your own models on ec2 or other gpu provider it doesn’t work