# Create high-performance kernels

Tilelang is a domain-specific language that facilitates the development of high-performance graphics processing unit (GPU), central processing unit (CPU), and accelerator kernels. This Python-based tool simplifies complex hardware optimization processes, allowing software developers to create more efficient computational units.

- ★ 8,208
- Python
- GitHub Trending · 2026-10-02

## Updates

- **October 2, 2026:** Stars 8,196 → 8,208, latest release v0.1.15 (September 30, 2026).

## What you get

- Creates fast computational units for GPU, CPU, and NPU
- Provides low-level optimization with Python-based syntax
- Offers high efficiency on the TVM compiler infrastructure

## Installation

**Standard installation**

```
pip install tilelang
```

**Verifying the installation**

```
python -c "import tilelang; print(tilelang.__version__)"
```

## If you don't write code

Paste this into your AI agent (Claude Code · Codex · Antigravity)

I want to develop a high-performance GPU kernel using Tilelang. Could you create a basic template for me to optimize operations like matrix multiplication or attention mechanisms using this Python-based structure? Could you explain the methods to increase efficiency, considering hardware optimizations and the TVM infrastructure?

Repository: https://github.com/tile-ai/tilelang. Follow the official instructions in the README of this repository for setup and execution.

Before starting, check which OS we are on and if the necessary tools (e.g., Python, Node.js, Git, Docker) are installed. Install any missing ones step-by-step according to this system.

Create a new, empty folder to work in. Do not touch any files outside this folder and do not delete anything. Tell me what you are going to do in one sentence before running each command.

Tell me what I should see on the screen at the end of each step. If an error occurs, I will paste the error message to you exactly as it is, and we will fix it together.

## Related dictionary terms

- [Domain-Specific Language](https://trescout.com/en/dictionary/domain-specific-language/)
- [Kernels](https://trescout.com/en/dictionary/kernels/)
- [CPU](https://trescout.com/en/dictionary/cpu/)
- [GPU](https://trescout.com/en/dictionary/gpu/)
- [Artificial Intelligence](https://trescout.com/en/dictionary/artificial-intelligence/)

- **Who it is for:** Intended for software engineers who want to develop high-performance computational units at the hardware level.

## Links

- [GitHub repository →](https://github.com/tile-ai/tilelang)
- [Read in Turkish →](https://trescout.com/discover/tilelang/)

TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-10-02: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. The introductory text was prepared with AI from the repository's own documentation, and this page was **machine-translated** from the Turkish original · the Turkish version prevails.

---
Source: TreScout Discover · https://trescout.com/en/discover/tilelang/