fix(ai): stop giving up on a model that is still working - #75
Open
arthware-dev wants to merge 2 commits into
Open
fix(ai): stop giving up on a model that is still working#75arthware-dev wants to merge 2 commits into
arthware-dev wants to merge 2 commits into
Conversation
The AI client waited for a whole response with nothing on the wire, so the only question it could ask was whether too much time had passed. A long document's prefill looks exactly like a stuck server from there, so filing a big scan on a small Mac got cancelled mid-flight and every minute already spent on it was thrown away. It now streams and times the silence instead. While an answer keeps arriving there is no deadline at all, however slow the machine. A server that actually stops is caught in 90 seconds rather than 300.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stacked on #74, which raises the ceilings. This changes what the timeout
measures, so the ceilings stop mattering.
The client waited for a whole response with nothing on the wire, so the
only question it could ask was whether too much time had passed — and a
long prefill looks exactly like a stuck server from there. It now streams
and times the silence instead: while an answer keeps arriving there is no
deadline, and a server that actually stops is caught in 90s rather than
900s.
Merge #74 first.