Repository navigation
Withdrawn - #1063
Closed
philip-pentatonic wants to merge 1 commit into
Closed
Withdrawn#1063philip-pentatonic wants to merge 1 commit into
philip-pentatonic wants to merge 1 commit into
Conversation
…ide a changing system message reuses nothing The CUDA engine keeps `TF_KEEP_MANY` prompt states (8 on the box) and takes them at the second message's start, the last assistant start and the prompt end: no grid, no marker, no chosen position, and a stored DeltaNet state cannot be truncated. Measured against production (17 requests): the same first message and a different user text reuse 1,276 of 1,288 tokens (223 to 38 ms), a first message that shares its whole leading block and differs after it reuses 0 of 1,297 (209 ms), and a twice-used block survives eight single-use prompts. So chat's 0 per cent is the shape of the system message, not the pool: a byte-identical rules message with the tools after it puts the resume point at the end of the stable block, and keeping the pool hot needs `TF_KEEP_MANY` 10 and the tool list fixed. The MLX prompt-cache flags (`--prompt-cache-gib`, `--checkpoint-slots`, `--spill-gib`, `--snapshot-dir`) are never forwarded on CUDA, so this engine has no byte budget and no disk persistence.
philip-pentatonic
deleted the
see-5083-tensorfold-prefix-cache-retention-keep
branch
October 19, 2026 09:58
philip-pentatonic
changed the title
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Withdrawn: opened in error; nothing is proposed upstream.