Conversation
Raise the memory-based default threshold from 24GB to 64GB (2048 below, 4096 at/above), and drop the inaccurate "128 for MoE" hint from the --chunk-size help. Benchmarks show larger chunks help prefill on both dense and MoE models, so the knob is available memory, not architecture.
stikves
marked this pull request as ready for review
September 22, 2026 19:38
kevchengcodes
approved these changes
Sep 22, 2026
chpriyal
approved these changes
Sep 22, 2026
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Raise the memory-based default threshold from 24GB to 64GB (2048 below, 4096 at/above), and drop the inaccurate "128 for MoE" hint from the
--chunk-sizehelp. Benchmarks show larger chunks help prefill on both dense and MoE models, so the knob is available memory, not architecture.