Llm
vllm-project/vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
vllm-project/vllm is an open-source project in the llm space, written mainly in Python.
It sits at around 91k stars and 21.7k forks, and was last pushed on 2026-09-04.
We keep this entry in sync on a schedule, so the counts and the description track the upstream project.
| Category | Llm |
|---|---|
| Stars | 91k |
| Forks | 21.7k |
| Language | Python |
| Last pushed | 2026-09-04 |
| Topics | amd, blackwell, cuda, deepseek |
git clone https://github.com/vllm-project/vllm.gitHow to use it
- Clone the repository with the command above.
- Follow its own README for dependencies and setup.
- Check the open issues and last-push date before you build on it.
Compiled and written by HowToPrompts from public sources. Open on GitHub ↗
← Back to AI Repos