Fix modelbuilder discrepancy on benchmarking #226
Merged
Conversation
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
titaiwangms
changed the title
justinchuby
approved these changes
Sep 19, 2025
sdpython
approved these changes
Sep 20, 2025
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Outdated suggestions cannot be applied.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.
Modelbuilder has different ordering on past kv cache, so we could not get the correct match on results with pyTorch models. This PR addresses this with re-ordering function (fully annotated).
NOTE: I will have a follow up PR to simplify and legitimize
get_inputsto test the real world case of LLM: (1) prompt-processing (sequence length > 1 without KV cache) and (2) token generation (sequence length==1 with KV cache)