Skip to content

block_multihead_attention support V100 GQA #68104

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 9 commits into from
Sep 11, 2024

Conversation

zhink
Copy link
Contributor

@zhink zhink commented Sep 9, 2024

PR Category

Inference

PR Types

New features

Description

pcard-71500
PR67485基础上,block_multihead_attention支持GQA、MQA模型(非MHA模型)运行在Volta、Turing显卡上;综合上一个PR可支持block_multihead_attention全阶段运行在Volta、Turing显卡上

Copy link

paddle-bot bot commented Sep 9, 2024

你的PR提交成功,感谢你对开源项目的贡献!
请关注后续CI自动化测试结果,详情请参考Paddle-CI手册
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

@zhink zhink changed the title V100 block_multihead_attention支持V100显卡GQA模型 Sep 11, 2024
@zhink zhink changed the title block_multihead_attention支持V100显卡GQA模型 block_multihead_attention support V100 GQA Sep 11, 2024
@zhoutianzi666 zhoutianzi666 merged commit a601f00 into PaddlePaddle:develop Sep 11, 2024
29 of 30 checks passed
icpcccpc added a commit to icpcccpc/Paddle that referenced this pull request Oct 12, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants