Skip to content

[BE/docs] Add fp8 rowwise perf table to float8 training readme #2312

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.

Sign up for GitHub

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
danielvegamyhre merged 2 commits into from
Jun 6, 2025
Merged

[BE/docs] Add fp8 rowwise perf table to float8 training readme #2312

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
82bfaa5
add fp8 rowwise perf table
danielvegamyhre Jun 5, 2025
fe81d3f
add fp8 rowwise perf table to readme
danielvegamyhre Jun 5, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Binary file added docs/static/fp8-rowwise-perf.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
9 changes: 8 additions & 1 deletion torchao/float8/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -115,7 +115,9 @@ on using `torchao.float8` in a distributed setting.

# Performance

A common question about float8 training is "when is float8 linear faster vs bfloat16?". Given the M, K, N of the forward pass through your linear, you can reference the table below for a microbenchmark based speedup estimate on NVIDIA H100:
A common question about float8 training is "when is float8 linear faster vs bfloat16?". Given the M, K, N of the forward pass through your linear, you can reference the tables below for a microbenchmark based speedup estimate on NVIDIA H100:

### Tensorwise scaling

<img width="805" alt="float8_speedup" src="https://github.com/user-attachments/assets/5c5f2817-7eb7-4cab-bd03-49fe70cd31a8">

Expand All @@ -135,6 +137,11 @@ To reproduce the raw data for table above, you can run the following script
python benchmarks/float8/float8_roofline.py your_output_filename.csv --shape_gen_name sweep
```

### Rowwise scaling

<img width="805" alt="float8_rowwise_speedup" src="../../docs/static/fp8-rowwise-perf.png" />


## Derivation

In a bf16 linear, assume all of the time is spent in gemms. In a float8 linear, account for max_abs and casting overhead. We want to know when
Expand Down
Loading