
The vLLM project (PyTorch Foundation)
vLLM
A high-throughput server for running models on GPUs, built for many users at once.
- Licence
- Apache-2.0
- First released
- 2023 (3 years)
- Runs
- On your own machine or server
- Latest release
- 2 Nov 2026
The review
Measured against the charter
Facts as the project publishes them, with links to check each one.
Own it
- Open-source licenceApache-2.0 (official site, opens outside Firefly Collective)
- Runs on your own machine or serverYes
Secure it
- Publishes a security policyRead it (official site, opens outside Firefly Collective)
- Publishes its security fixesThe project's advisories (official site, opens outside Firefly Collective)
- A release in the last 12 months2 Nov 2026 (official site, opens outside Firefly Collective)
Make it last
- In use for five years or more2023 (3 years)
- Who maintains itThe vLLM project (PyTorch Foundation)
- Source code is publicBrowse the source (official site, opens outside Firefly Collective)
Facts as published by the project, checked 12 Nov 2026. A dash means not found or not applicable, not a judgement.
Firefly Collective is independent and not affiliated with or endorsed by the projects listed. Names and logos are the property of their owners. Summaries are provided as-is; always check the project’s official advisory.
Also in this kind
The other options
Ollama
Ollama, Inc. and contributors
llama.cpp
ggml.ai and contributors
LM Studio
Element Labs
vLLM
The vLLM project (PyTorch Foundation)
Think this list is missing something? Suggest it for the Stacks