Repository navigation
Conversation
Hugging Face selects the rows a logits_to_keep index tensor names as hidden[:, rows] over a [1, T, D] hidden state, which exports as aten.index.Tensor with a None before the index. The ggml engine gathered rows only of a matrix with one index tensor. It now takes that case too: over a leading size-1 dimension it gathers rows of the [T, D] matrix and keeps the leading 1. The indices are still checked on the host.
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/25008
Note: Links to docs will display an error until the docs builds have been completed. ❌ 137 New Failures, 7 Unrelated Failures, 10 Unclassified FailuresAs of commit c0d7b51 with merge base c7f7931 ( NEW FAILURES - The following jobs have failed:
UNCLASSIFIED FAILURES - DrCI could not classify the following jobs because the workflow did not run on the merge base. The failures may be pre-existing on trunk or introduced by this PR:
FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Nov 13, 2026
This branch had an error being deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Stack
Off-graph KV cache for the native ggml engine, through serving Muse Glimmer on it. Review bottom-up; each PR builds and passes its tests on its own.
Summary
Hugging Face selects the rows a
logits_to_keepindex tensor names ashidden[:, rows]over a[1, T, D]hidden state, which exports asaten.index.Tensorwith aNonebefore the index. The ggml engine gathered rows only of a matrix with one index tensor. It now takes that case too: over a leading size-1 dimension it gathers rows of the[T, D]matrix and keeps the leading 1. The indices are still checked on the host.Test plan
A new engine test gathers two rows of a
[1, 6, 8]input and rejects an out-of-range row. The C++ test harness gains an optional-tensor-list argument to build it.Authored with Claude Code.