Updated

vLLM

Open-source inference engine that serves open-weight models on GPUs behind an OpenAI-compatible API, with high throughput under concurrent load.

43 open-source projects declare vLLM in their code; no maker's live product we track uses it yet.

Open sourceSelf-hostable

Profile checked

Ask your AI about this, with this page as the source:ChatGPT ↗Claude ↗Perplexity ↗

How makers use vLLM

What vLLM is used for

Open Model Hosting & GPUs · Open-source serving engine →Serving open-weight models on your own or rented GPUs behind an OpenAI-compatible API, with high throughput for many concurrent requests.Trade-off: You own the GPUs, scaling, upgrades and on-call; tuning memory and batching settings takes real effort.
Hugging Face77Modal60Baseten23Runpod19Together AI12vLLM0Makers' products on each, for open model hosting & GPUs

vLLM vs the alternatives makers chose

OptionMakers' products2026 launchesOpen source
Hugging Face77 42%52%No
Modal60 32%48%No
Baseten23 12%2%No
Runpod19 10%2%No
Together AI12 6%0%No
Fireworks AI5 3%2%No
vLLM0 0%0%Yes

For open model hosting & GPUs: how many of the 185 makers' products we know the open model hosting & GPUs of chose each option, and the share among the 48 launched in 2026, since a tool that came first has products from every year.

When to use vLLM, and when not

Pick vLLM if

Something else fits better if

  • The open model you need is already served by a host and you want to pay per token: Together AI or Fireworks AI or Hugging Face.
  • Your workload is your own Python code — a pipeline, batch job or custom model — that should scale to zero: Modal.
  • You want a raw GPU machine for experiments, training or a long-running server: Runpod.
  • One model is core to the product and needs steady latency under real traffic: Baseten or Hugging Face.
From each decision's guide, written from what makers' products chose and why.

How vLLM itself is built

2 tools, from its own code, website and Product Hunt page.

Who uses it

No maker's live product on record yet, and 43 open-source projects that declare it in their code.

We haven't found a maker's live product that uses vLLM yet — only the open-source projects below, which declare it in their code.

Daft
DaftHigh-performance data engine for AI and multimodal workloads. Process images, audio, vid…Open Model Hosting & GPUsIn its code · source ↗
pruna
prunaPruna is a model optimization framework built for developers, enabling you to deliver fa…Open Model Hosting & GPUsIn its code · source ↗

Open source: a project that declares vLLM as a dependency in its public code — verifiable, but not necessarily a live product.

What its projects tell coding agents

10 of the 26 open-source products built with vLLM that were worked on in the last 180 days keep instruction files for coding agents — 38%, against 53% of all the projects we read. AGENTS.md is in 10, CLAUDE.md in 6.

880words in the median file
uvthe package manager most named, in 4 of 10 files
70%say how to test a change
Named more in these files than in other projects' — in how many of the 10:
What AGENTS.md files say about other tools →What open-source products tell their coding agents →

Questions makers ask about vLLM

What are the alternatives to vLLM?

For open model hosting & GPUs, the other options makers' products we track chose most are Hugging Face (77), Modal (60), Baseten (23), Runpod (19), Together AI (12), Fireworks AI (5). Open source or self-hostable among them: Baseten. Each alternative, with the situation it fits, is on the vLLM alternatives page.

When should you use vLLM?

You run the GPUs yourself and need an OpenAI-compatible server for many concurrent users (open model hosting & GPUs). Something else fits better if: the open model you need is already served by a host and you want to pay per token, pick Together AI or Fireworks AI or Hugging Face; your workload is your own Python code — a pipeline, batch job or custom model — that should scale to zero, pick Modal.

Is vLLM open source or self-hostable?

Open source, and you can self-host it. source ↗

Can AI coding agents work with vLLM?

No llms.txt, official MCP server or CLI found yet.

Who uses vLLM?

No maker's live product we track yet; 43 open-source projects declare it in their code. source ↗