Dao-AILab/flash-attention
Dao-AILab/flash-attention has 107,405 source lines of code out of 135,654 total lines across 996 files, counted from the main ref at commit 1326ab1ddf18 by the OctoCounts tokei engine on 2026-10-22. The report is cached by commit, tokei version, and analysis options, so recounting the same revision returns exactly these numbers.
As of 2026-10-22 (commit 1326ab1ddf18), Dao-AILab/flash-attention contains 135,654 total lines: 107,405 code, 15,096 comments, 13,153 blank, across 996 files in 13 languages (top: Python 74.9%). Custom configuration (tests excluded, docs excluded, generated excluded). Counted with tokei via OctoCounts.
OctoCounts produced this report by resolving Dao-AILab/flash-attention to commit 1326ab1ddf18, 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 79.2% of all lines, comments represent 11.1%, and the repository averages 108 code lines per file. Python accounts for 74.9% of counted code.
| Language | Files | Lines | Code | Comments | Blanks |
|---|---|---|---|---|---|
| Python | 181 | 95798 | 80455 | 7506 | 7837 |
| C Header | 43 | 11552 | 8571 | 1553 | 1428 |
| C++ | 12 | 9001 | 7492 | 611 | 898 |
| C++ Header | 12 | 6780 | 5406 | 688 | 686 |
| CUDA | 606 | 7453 | 4199 | 1940 | 1314 |
| YAML | 107 | 1440 | 1042 | 209 | 189 |
| Shell | 5 | 139 | 84 | 38 | 17 |
| TOML | 2 | 82 | 72 | 0 | 10 |
| 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 74.9%). Generated at .
Reproduce this report
Report FAQ
How many lines of code does Dao-AILab/flash-attention have?
Dao-AILab/flash-attention has 135,654 total lines, including 107,405 code lines, 15,096 comment lines, and 13,153 blank lines.
How was the Dao-AILab/flash-attention line count measured?
OctoCounts resolved the public GitHub repository to commit 1326ab1ddf18, 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 1326ab1ddf18 on 2026-10-22T05:35:05.022628585+00:00.
Similar repository reports
Other public repositories with an OctoCounts report, ranked by top language and code size similarity:
- python-pillow/Pillow — Python, 105,753 code lines
- SanjeevBashyal/planetx — Python, 105,359 code lines
- MSKazemi/kubeintellect — Python, 104,462 code lines
- ifixai-ai/iFixAi — Python, 110,450 code lines
- microsoft/autogen — Python, 111,060 code lines
- context-labs/HALO — Python, 103,313 code lines
Play the growth animation for Dao-AILab/flash-attention on this page — 107,405 code lines across 12 languages, counted by tokei.