Source: https://seerepo.com/compare/Dao-AILab/flash-attention/vs/tile-ai/tilelang
Markdown source: https://seerepo.com/compare/Dao-AILab/flash-attention/vs/tile-ai/tilelang.md
Description: flash-attention and tilelang share 5 technologies: apache-tvm-ffi, NVIDIA, torch-c-dlpack-ext, torch, and einops. Both are written in Python.

# flash-attention vs tilelang

flash-attention and tilelang share 5 technologies: apache-tvm-ffi, NVIDIA, torch-c-dlpack-ext, torch, and einops. Both are written in Python.

- [Dao-AILab/flash-attention](https://seerepo.com/r/Dao-AILab/flash-attention): Fast and memory-efficient exact attention
- [tile-ai/tilelang](https://seerepo.com/r/tile-ai/tilelang):  Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

## What they share and what sets them apart

- In common: apache-tvm-ffi, NVIDIA, torch-c-dlpack-ext, torch, einops
- Only in flash-attention: quack-kernels, triton-windows, triton
- Only in tilelang: tornado, seaborn, matplotlib, cuda-python, flash-attn, flash-linear-attention, ml-dtypes, docutils, psutil

## Side by side

| | flash-attention | tilelang |
|---|---|---|
| Size | 139,521 lines written, 150 code files | 324,472 lines written, 455 code files |
| Who writes it | 1.7% of its commits signed by AI. With Claude Code, Cursor, and GitHub Copilot | 4.5% of its commits signed by AI. With Claude Code, CodeRabbit, GitHub Copilot, Gemini CLI, Codex, and Cursor |
| Activity | Last push November 16, 2026, 1,613 commits, 25,304 stars | Last push November 13, 2026, 2,194 commits, 8,976 stars |
| Tests | 80 test files. 32% of its code files have a test that imports them | 633 test files. 37.6% of its code files have a test that imports them |

- Size: Lines written by people and agents, without generated ones; code files, without tests.
- Who writes it: A commit counts as AI's when it carries its signature: a Co-Authored-By line, its author name or the merge of its branch.
- Activity: From GitHub and its history, on the day SeeRepo read it.
- Tests: The files that test the code, and the share of code files a test imports.

## Frequently asked questions

### Which is bigger, flash-attention or tilelang?

tilelang, with 324,472 lines written; flash-attention has 139,521.

The proof: [tile-ai/tilelang's code at commit 210611d](https://github.com/tile-ai/tilelang/tree/210611dca01198cacac9a02103c2a95e18cf6375)

### Which has more GitHub stars, flash-attention or tilelang?

flash-attention, with 25,304 stars; tilelang has 8,976.

The proof: [Dao-AILab/flash-attention on GitHub](https://github.com/Dao-AILab/flash-attention)

### Which has more tests, flash-attention or tilelang?

tilelang, with 633 test files; flash-attention has 80.

The proof: [tile-ai/tilelang's code at commit 210611d](https://github.com/tile-ai/tilelang/tree/210611dca01198cacac9a02103c2a95e18cf6375)

### Which was updated more recently, flash-attention or tilelang?

flash-attention: its last push was on November 16, 2026, and tilelang's on November 13, 2026.

The proof: [Dao-AILab/flash-attention's commits on GitHub](https://github.com/Dao-AILab/flash-attention/commits)

### Which has more commits signed by AI, flash-attention or tilelang?

tilelang: an AI signed 4.5% of its commits, with Claude Code, CodeRabbit, GitHub Copilot, Gemini CLI, Codex, and Cursor; in flash-attention, 1.7%.

The proof: [the commits each tool signed in tile-ai/tilelang](https://seerepo.com/r/tile-ai/tilelang/ai)

## Sources

- Dao-AILab/flash-attention: Read on November 16, 2026, at commit 328dd7b (https://github.com/Dao-AILab/flash-attention/tree/328dd7b5a6905c75500a9bce334ac012763409a9)
- tile-ai/tilelang: Read on November 15, 2026, at commit 210611d (https://github.com/tile-ai/tilelang/tree/210611dca01198cacac9a02103c2a95e18cf6375)

## What is your repo built with?

Any public repo, with no account: https://seerepo.com/r/owner/repo/built, with the owner and name it has on GitHub.
