itowlson

@codeitlikemiley Yes, local is the default. The guide for setting up local LLMs is at https://spinframework.dev/v3/serverless-ai-api-guide#file-structure. _Please allow plenty of testing time._ Local LLM is not a feature that gets...

The only thing I can suggest offhand is to write your own k8s service that implements the `cloud-gpu` HTTP API (https://github.com/fermyon/spin-cloud-gpu/blob/main/fermyon-cloud-gpu/src/index.ts#L57 ff), then point at that in your `runtime-config.toml` (or...

+1 for build something first

We are now back on official async-tar but the broader issue about async-std remains.

A separate tutorial seems reasonable during the transitional period (then make a call when P3 has sufficient adoption to retire P2). It would also be good to get stuff in...

@lann We added some checks to the workflow to improve parity with main CI, but they are not 'required' (because they were added after the 'required' stuff got set up)....

@rylev and SDK folks: do you know if/how: > HttpRequestDenied used to be returned from wasi:http/outgoing-handler#handle but are now returned from wasi:http/types/future-incoming-response#get would affect developers using the SDKs, as opposed...

@rylev re "Change how incoming-request.authority is set #2723" - could you expand on what changes in the observable behaviour please? The "moral equivalent" #2684 says "The guest will now see...

Thanks for the confirmation re the HTTP error. I'll double check other SDKs but I think they all hide this (and yep, understood for folks working at the WASI level)....

Aha! This is the missing piece. Let me word it back to check I now have it right: > Spin now infers the `authority` property of an incoming HTTP request...