Home Projects ZhiLight
ZhiLight
C++

ZhiLight

A highly optimized LLM inference acceleration engine for Llama and its variants.

by zhihu · GitHub
Stars
Forks
License
Created
Last commit
Category
Language
Likely
Self-hostable
cudadeepseek-r1gptApache-2.0C++Self-hostable
View on GitHub
In plain words

Speed up the performance of Llama AI models on compatible graphics cards for faster results.

You maintain this project?

Claim its page: indexed whatever its rank, translated into six languages, and enriched with what you write yourself.

Claim this page →
ZhiLight — GitHub preview card
📈 Star history
906905
2026-07-202026-08-31
📈 Track ZhiLight

Get an email alert on its next release or when it starts trending — never miss the moment.

Free · no card · unsubscribe anytime
Get email alerts →
📄 About

A highly optimized LLM inference acceleration engine for Llama and its variants.

ZhiLight has 905 stars on GitHub. It has been forked 103 times. ZhiLight is written mainly in C++. It has been in active development since 2024. ZhiLight is available under the Apache-2.0 license. Its main topics are cuda, deepseek-r1, gpt, inference-engine.

Frequently asked questions

What is ZhiLight?

A highly optimized LLM inference acceleration engine for Llama and its variants.

Is ZhiLight open source?

ZhiLight is an open-source project. It is released under the Apache-2.0 license.

Is ZhiLight free?

Yes. ZhiLight is free and open source — you can use, modify and self-host it.

What license does ZhiLight use?

ZhiLight is available under the Apache-2.0 license.

What language is ZhiLight written in?

ZhiLight is written mainly in C++.

🏅 Maintainer of this project?
olud.ai badge — ZhiLight

Add this live badge to your README — your GitHub stars and directory rank, refreshed daily.

[![olud.ai](https://olud.ai/badge.php?tool=zhihu-zhilight)](https://olud.ai/project/zhihu-zhilight.html)
More badge options →
🧬 Shares DNA with🧬 View the DNA map →

Measured from GitHub topics shared by both projects, weighted by how rare each topic is.