Skip to content
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

Add flash_attention_benchmark and gemm_benchmark #83

Closed
wants to merge 11 commits into from
Closed

Conversation

xuzhao9
Copy link
Contributor

@xuzhao9 xuzhao9 commented Nov 29, 2024

Add flash_attention_benchmark and gemm_benchmark.
Right now, only run triton_tutorial_flash_attn_v2 backend with a single input (4, 32, 16384, 64) with backward pass.

Test:

# Install Tritonbench library
$ pip install -e .
# Run benchmark script
$ python benchmarks/flash_attention_bench/run.py
  (Batch, Heads, SeqLen, Dhead)    triton_tutorial_flash_v2-tflops
-------------------------------  ---------------------------------
             (4, 32, 16384, 64)                            264.718

@xuzhao9 xuzhao9 changed the title Add flash_attention_benchmark Add flash_attention_benchmark and gemm_benchmark Dec 4, 2024
@xuzhao9 xuzhao9 marked this pull request as draft December 4, 2024 16:40
@xuzhao9 xuzhao9 temporarily deployed to docker-s3-upload December 4, 2024 23:45 — with GitHub Actions Inactive
@xuzhao9 xuzhao9 temporarily deployed to docker-s3-upload December 4, 2024 23:45 — with GitHub Actions Inactive
@xuzhao9 xuzhao9 marked this pull request as ready for review December 5, 2024 00:55
@xuzhao9 xuzhao9 requested review from FindHao and adamomainz December 5, 2024 00:55
@facebook-github-bot
Copy link
Contributor

@xuzhao9 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

Copy link
Member

@FindHao FindHao left a comment

Choose a reason for hiding this comment

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

LGTM!

@facebook-github-bot
Copy link
Contributor

@xuzhao9 merged this pull request in e7c7d99.

@xuzhao9 xuzhao9 deleted the xz9/add-gemm branch December 5, 2024 18:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants