# Dao-AILab/flash-attention SLOC report

> As of 2026-09-18 (commit 1bda8f9290cd), Dao-AILab/flash-attention contains 159,860 total lines: 129,194 code, 14,946 comments, 15,720 blank, across 1,037 files in 13 languages (top: Python 79.2%). Default configuration. Counted with tokei via OctoCounts.

Dao-AILab/flash-attention has 129,194 source lines of code out of 159,860 total lines across 1,037 files, counted from the main ref at commit 1bda8f9290cd by the OctoCounts tokei engine on 2026-09-18. The report is cached by commit, tokei version, and analysis options, so recounting the same revision returns exactly these numbers.

OctoCounts produced this report by resolving Dao-AILab/flash-attention to commit 1bda8f9290cd, downloading the repository source archive, and counting every source file with tokei, the open-source line counter written in Rust. The table below breaks the count down by programming language into files, total lines, code lines, comment lines, and blank lines, so the figures can be compared across languages and projects. Results are cached by commit, tokei version, and analysis options, so counting the same revision again reproduces exactly these numbers.

## Repository size insights

This is a large codebase by counted code lines. Code represents 80.8% of all lines, comments represent 9.3%, and the repository averages 125 code lines per file. Python accounts for 79.2% of counted code.

| Language | Files | Lines | Code | Comments | Blanks |
| --- | ---: | ---: | ---: | ---: | ---: |
| Python | 226 | 121,027 | 102,288 | 8,174 | 10,565 |
| C Header | 43 | 11,552 | 8,571 | 1,553 | 1,428 |
| C++ | 12 | 8,979 | 7,471 | 610 | 898 |
| C++ Header | 12 | 6,773 | 5,399 | 688 | 686 |
| CUDA | 606 | 7,453 | 4,199 | 1,940 | 1,314 |
| YAML | 107 | 1,440 | 1,042 | 209 | 189 |
| TOML | 3 | 85 | 75 | 0 | 10 |
| Shell | 4 | 104 | 65 | 22 | 17 |
| Dockerfile | 2 | 122 | 56 | 46 | 20 |
| Autoconf | 2 | 17 | 16 | 0 | 1 |
| Makefile | 1 | 9 | 6 | 0 | 3 |
| Module-Definition | 1 | 7 | 6 | 0 | 1 |

Top language (Python 79.2%). Generated at 2026-09-18T23:56:54.078339754+00:00. Canonical report: https://octocounts.com/github/Dao-AILab/flash-attention

## Reproduce this report

[Reproduce this exact report and configuration](https://octocounts.com/github/Dao-AILab/flash-attention/commit/1bda8f9290cd48d030f1516f0e680cd464ef3554?analysis=%7B%22ignoredDirs%22%3A%5B%5D%2C%22ignoredLanguages%22%3A%5B%5D%2C%22profile%22%3A%22default%22%2C%22includeDocs%22%3Atrue%2C%22includeTests%22%3Atrue%2C%22includeGenerated%22%3Atrue%7D).

## Report FAQ

### How many lines of code does Dao-AILab/flash-attention have?

Dao-AILab/flash-attention has 159,860 total lines, including 129,194 code lines, 14,946 comment lines, and 15,720 blank lines.

### How was the Dao-AILab/flash-attention line count measured?

OctoCounts resolved the public GitHub repository to commit 1bda8f9290cd, downloaded the source archive, counted it with tokei, and cached the report by commit, tokei version, and analysis options.

### What commit was counted for Dao-AILab/flash-attention?

This OctoCounts report was generated from main at commit 1bda8f9290cd on 2026-09-18T23:56:54.078339754+00:00.

## Similar repository reports

- [ansible/ansible](https://octocounts.com/github/ansible/ansible) — Python, 127,514 code lines
- [fla-org/flash-linear-attention](https://octocounts.com/github/fla-org/flash-linear-attention) — Python, 132,071 code lines
- [InnerAura/hyperweave](https://octocounts.com/github/InnerAura/hyperweave) — Python, 135,436 code lines
- [pypa/pip](https://octocounts.com/github/pypa/pip) — Python, 120,992 code lines
- [fastapi/fastapi](https://octocounts.com/github/fastapi/fastapi) — Python, 117,708 code lines
- [trpc-group/trpc-agent-python](https://octocounts.com/github/trpc-group/trpc-agent-python) — Python, 141,078 code lines

## Related OctoCounts pages

- [Recently analyzed repositories](https://octocounts.com/recent)
- [Popular SLOC reports](https://octocounts.com/popular)
- [Trending GitHub repositories](https://octocounts.com/trending)
- [Hall of Monoliths](https://octocounts.com/hall-of-monoliths)
- [TensorFlow vs PyTorch](https://octocounts.com/compare/tensorflow-vs-pytorch)
- [Django vs Ruby on Rails](https://octocounts.com/compare/django-vs-rails)
- [Laravel vs Django](https://octocounts.com/compare/laravel-vs-django)
- [GitHub SLOC counter guide](https://octocounts.com/docs/github-sloc-counter)
- [Counting methodology](https://octocounts.com/docs/methodology)
- [SLOC and code metrics glossary](https://octocounts.com/docs/glossary)
- [Original research: how filtering changes SLOC counts](https://octocounts.com/research)
- [OctoCounts API docs](https://octocounts.com/docs/api)