Add batching of streamed chat response
What does this MR do and why?
Add batching of streamed chat response
This adds a batched response for streamed chat responses by only sending every second token as a batched response.
As we have a significant overhead for each token via our GraphQL subscription JSON response, this should at least reduce the amount of websocket messages we need to send for each chat response.
Changelog: changed EE: true
MR acceptance checklist
Please evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.
Screenshots or screen recordings
Screenshots are required for UI changes, and strongly recommended for all other merge requests.
Before | After |
---|---|
How to set up and validate locally
Numbered steps to set up and validate the change are strongly suggested.