Repository navigation
Update Gemma 4 benchmarks with static INT2 SME2 kernels - #3943
Merged
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Update the Gemma 4 benchmark to use tested LiteRT-LM, KleidiAI 1.31, and XNNPACK revisions with a matching baseline. The baseline already includes earlier SME2 support, so the results isolate the additional static INT2 kernels.
Keep the setup focused on macOS and add one comparison table for Apple M4 Pro and Android. Explain thread-count trade-offs, variability, and regressions, with a compact Android Notice linking existing setup guidance and the SME2 device list.
Validation:
Hugo build: 4,866 pages, with no warnings or errors.
Front matter, page ordering, internal anchors, conflict-marker scan, and all 15 Bash blocks pass.
Previously completed hardware validation for these source revisions includes one- and four-thread benchmarks, prompt checks, profiling, and 45 targeted Android kernel, operator, and subgraph tests.
I have reviewed Create a Learning Path.
I have checked my contribution for confidential information.
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of the Creative Commons Attribution 4.0 International License.