Skip to content

Support bf16 activations in Llama4 ExecuTorch XNNPACK export (#21784) - #21784

Open
JakeStevens wants to merge 1 commit into
pytorch:mainfrom
JakeStevens:export-D115569455
Open

Support bf16 activations in Llama4 ExecuTorch XNNPACK export (#21784)#21784
JakeStevens wants to merge 1 commit into
pytorch:mainfrom
JakeStevens:export-D115569455

Conversation

@JakeStevens

@JakeStevens JakeStevens commented Aug 12, 2026 •
edited by meta-codesync Bot
Loading

Copy link
Copy Markdown
Contributor

Summary:

Adds a working --precision bf16 path to export_llm_backbone so a Llama4 mini checkpoint can be exported with bf16 activations instead of the default fp32 activations, while keeping the quantized linears delegated to XNNPACK.

Reviewed By: billmguo, telgamal-1

Differential Revision: D115569455

@pytorch-bot

pytorch-bot Bot commented Aug 12, 2026 •
edited
Loading

Copy link
Copy Markdown

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21784

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 18c2a62 with merge base 3b5bf0f (image):
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@meta-cla meta-cla Bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Aug 12, 2026
@meta-codesync

meta-codesync Bot commented Aug 12, 2026

Copy link
Copy Markdown
Contributor

@JakeStevens has exported this pull request. If you are a Meta employee, you can view the originating Diff in D115569455.

@github-actions

github-actions Bot commented Aug 12, 2026

Copy link
Copy Markdown

This PR needs a release notes: label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

@JakeStevens @facebook-github-bot
Support bf16 activations in Llama4 ExecuTorch XNNPACK export (pytorch…
18c2a62 
…#21784)

Summary:

Adds a working `--precision bf16` path to `export_llm_backbone` so a Llama4 mini checkpoint can be exported with bf16 activations instead of the default fp32 activations, while keeping the quantized linears delegated to XNNPACK.

Reviewed By: billmguo, telgamal-1

Differential Revision: D115569455
@meta-codesync meta-codesync Bot changed the title Support bf16 activations in Llama4 ExecuTorch XNNPACK export Support bf16 activations in Llama4 ExecuTorch XNNPACK export (#21784) Aug 12, 2026
digantdesai
digantdesai approved these changes Aug 12, 2026
View reviewed changes

@digantdesai digantdesai left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Review automatically exported from Phabricator review in Meta.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@digantdesai digantdesai digantdesai approved these changes
@larryliu0820 larryliu0820 Awaiting requested review from larryliu0820 larryliu0820 is a code owner
@mergennachin mergennachin Awaiting requested review from mergennachin mergennachin is a code owner

Assignees

No one assigned

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. meta-exported

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@JakeStevens @digantdesai