vLLM
by vLLM Project
At a glance
Open-source inference and serving engine for large language models, known for high-throughput PagedAttention.
open-source inference serving llm
- Licence
- Apache-2.0
- Category
- Platforms & Infrastructure
- Business model
- Open (open weights/source)
- Availability
- Global
- Launched
- 2023-06
Overview
vLLM is an open-source inference and serving engine for large language models, maintained by the vLLM Project and originally developed at UC Berkeley, whose PagedAttention memory management underpins its throughput. It is software you install and run yourself rather than a vendor-hosted service, and it ships an OpenAI-compatible API server, with API and CLI references in its documentation. No material change was found in this review. The main limitation is pricing and region: vLLM publishes no vendor price list, since it is self-hosted open-source software whose cost is the GPU hardware you run it on, and regional coverage is unstated in the record.
Compare this tool Suggest a correction
Key features
- PagedAttention for efficient KV-cache memory management
- High-throughput continuous batching of requests
- OpenAI-compatible API server out of the box
- Broad support for open model architectures and quantization
Use cases
- Serving open LLMs in production at high throughput
- Self-hosting an OpenAI-compatible inference endpoint
- Benchmarking and research on efficient LLM inference
Pricing
vLLM is free and open source under Apache-2.0; costs are limited to the GPU infrastructure it runs on.
Record details
Provenance, dates and links are kept here so the main sections stay about the product itself.
Links, dates and provenance
- Record updated
- 2026-10-10
- Documentation
- docs.vllm.ai ↗
- Founded
- 2023
- Headquarters
- US
- Canonical URL
- https://globalaiproductindex.com/products/vllm/
Selection profile
Every field says where its value comes from. "Unknown" means we have not confirmed it — it does not mean the product lacks the feature.
Selection profile11 fields4 unknown
- Tasks it fits
- Running models yourself
Derived from the record: category "platforms".
Suggest a correction - How you use it
- Self-run service
Open-source project without a tag describing its delivery form; run it yourself.
Suggest a correction - Runs on
- Not recorded
No platform support is stated in the record.
Suggest a correction - Where it runs
- Not recorded
Open-source record without a hosting signal; deployment target not recorded.
Suggest a correction - Works offline
- Not recorded
The record contains no offline-operation claim.
Suggest a correction - Official service region
- Global
Mirrors the availability field already on the record ("global").
Suggest a correction - Public API
- Yes
Matched on the vendor page: “itializing search GitHub Home User Guide Developer Guide Benchmarking API Reference CLI Reference Community vLLM GitHub Home User Guide User Guide Getting Started Getting”
Suggest a correction - Licence
- Apache-2.0
The repository licence endpoint returns Apache-2.0.
Suggest a correction - Commercial use
- Permitted
Recorded licence allows commercial use (copyleft obligations may still apply).
Suggest a correction - Interface languages
- Not recorded
No vendor page read in this pass states the product's own interface languages; the key exists so the field can be answered later rather than left absent.
Suggest a correction - Billing
- Open source Free Free allowance: Yes Subscription required: No
Prices are the vendor’s advertised entry price and exclude taxes; always confirm on the vendor’s own page.
Suggest a correction
Where each value comes from
- Tasks it fits
- Our inference
- How you use it
- Our inference
- Runs on
- Our inference
- Where it runs
- Our inference
- Works offline
- Our inference
- Official service region
- Our inference
- Public API
- Verified 2026-10-09 docs.vllm.ai ↗
- Licence
- Verified 2026-10-10 github.com ↗
- Commercial use
- Our inference
- Interface languages
- Unknown
Key facts and their evidence
Each fact lists the conditions it holds under, the page it was read from and the day it was checked.
-
API: Public API documented Verified 2026-10-09 docs.vllm.ai ↗
Applies to: As stated on docs.vllm.ai on 2026-10-09
Matched on the vendor page: “itializing search GitHub Home User Guide Developer Guide Benchmarking API Reference CLI Reference Community vLLM GitHub Home User Guide User Guide Getting Started Getting”
Suggest a correction -
Licence: Apache-2.0 Verified 2026-10-10 github.com ↗
Suggest a correction
Frequently asked questions
What is vLLM used for?
vLLM serves large language models with high throughput, exposing an OpenAI-compatible API for self-hosted deployments.
Is vLLM free?
Yes. vLLM is open source under Apache-2.0; you only pay for the hardware you run it on.
What is PagedAttention?
PagedAttention is vLLM's memory-management technique that pages the KV cache, raising GPU utilization and throughput.
Who should not choose vLLM?
Teams that want a managed API endpoint rather than running their own. vLLM is open-source software you self-host under Apache-2.0, the record lists no vendor price list, and no hosted service is recorded.
Similar products
- LM Studio — Desktop application for discovering, downloading, and running open large language models locally.
- Ollama — Open-source tool for running open large language models locally on your own machine with a simple command-line interface.
- Jan — Open-source desktop app for running open language models locally as an offline ChatGPT alternative.
- Open WebUI — Extensible, self-hosted AI interface that runs offline and connects to local or remote models such as Ollama and OpenAI-compatible APIs.
- Baseten — Platform for deploying and serving machine-learning models as scalable APIs.
- Cerebras — AI compute company whose inference cloud serves open language models on wafer-scale hardware.
All vLLM alternatives → · All Platforms & Infrastructure products →
Related guides
Sources
This record was last reviewed on 2026-10-10.
Machine-readable record: /api/products/vllm.json · Spot an error? Suggest a correction