Repository navigation
Conversation
…n history Text-based multi-turn callers re-render history without control tokens, so an SR turn's header renders its site token instead of the control token that was sent and the prefix cache misses from there on. history_adapters (one entry per assistant message) restores them in place of the site token, for both the 4.1 role-marker and 4.2 ChatML templates. Adds tests and documents the option and the caller rules it needs in SR_ARCHITECTURE.md. Signed-off-by: noaa <noaa.kless@gmail.com>
noaakl
requested review from
antonpibm,
aviv1ron1,
freunda and
yairallouche
as code owners
October 25, 2026 16:46
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Add history_adapters option to keep SR control tokens in history
Text callers re-render history without control tokens, so a past SR turn
shows its site token instead of the control token that was sent. The ids
stop matching and the prefix cache misses from that point on.
history_adapters lists the adapter behind each assistant message, and the
chat template puts each control token back in place of the site token.
Works for both the Granite 4.1 and 4.2 chat templates.