-
Notifications
You must be signed in to change notification settings - Fork 77
Update the muxing rules to v3 #1112
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
Marking as draft because this would break the FE. We need to coordinate to merge |
API Version 3? |
on it stacklok/codegate-ui#319 👨💻 |
uhm... I'd be careful with having a v3 that's not backwards compatible without more usecases. This already starts looking too optimized for fim/chat in a way that it might not cater for other rule types. |
As I see it, it will be optimized for the request type. Right now we can only distinguish between 2 request types, fim or chat. Hence the types we have. In the future we would be able to match more request types, e.g. request type: Persona based. But I was afraid of the same thing, of the types being very restrictive. In the end I convince myself that it wasn't as restrictive as I initially thought. But of course all discussions welcomed and any feedback is appreciated 😃 |
Let's see this in action and I'll keep reviewing in the meantime. |
0edea97
to
0890c1a
Compare
9ea1504
to
d5957e0
Compare
Closes: #1060 Right now the muxing rules are designed to catch globally FIM or Chat requests. This PR extends its functionality to be able to match per file and request, i.e. this PR enables - Chat request of main.py -> model 1 - FIM request of main.py -> model 2 - Any type of v1.py -> model 3
d5957e0
to
34f4f02
Compare
Closes: #1060
Right now the muxing rules are designed to catch globally FIM or Chat requests. This PR extends its functionality to be able to match per file and request, i.e. this PR enables