fix(llm): include system prompt tokens in memory compressor budget#3
Draft
fix(llm): include system prompt tokens in memory compressor budget#3
Conversation
0xhis
force-pushed
the
fix/memory-compressor-token-budget
branch
from
2502a59 to
22d978e
Compare
The memory compressor was not accounting for system prompt and agent identity tokens when calculating the conversation budget. This caused premature history truncation on long scans with large system prompts. Adds a reserved_tokens parameter to compress_history() that subtracts already-accounted tokens from the budget before applying limits.
0xhis
force-pushed
the
fix/memory-compressor-token-budget
branch
from
22d978e to
63035db
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
The memory compressor was not accounting for system prompt and agent identity tokens when calculating the conversation budget. This caused premature history truncation on long scans with large system prompts.
Changes
reserved_tokensparameter tocompress_history()that subtracts already-accounted tokens from the budget before applying limitsreserved_tokensfrom system prompt and agent identity messages in_prepare_messages()Files Changed
strix/llm/llm.py(+7/-2)strix/llm/memory_compressor.py(+8/-1)Split from usestrix#328.