vLLM
High-throughput LLM serving engine.
Open-source inference engine with PagedAttention and continuous batching; the de-facto standard for self-hosted serving.
- Vendor
- vLLM project
- Category
- Inference & model hosting
- Pricing
- Open source
- Open source
- Yes
- License
- Apache 2.0
- Platforms
- Linux, GPU, TPU
- Launched
- 2023
- Website
- vllm.ai
Features
- PagedAttention
- Continuous batching
- OpenAI-compatible server
Best for
- Local & on-device
- High volume / low cost
More inference & model hosting
- Ollama — Ollama. Run open models locally with one command.
- OpenRouter — OpenRouter. One API for hundreds of models.
- llama.cpp — ggml. LLM inference in C/C++ on any hardware.
- Amazon Bedrock — AWS. Foundation models on AWS.
- Vertex AI — Google Cloud. Google Cloud’s AI platform.
- LM Studio — Element Labs. Desktop app to discover and run local LLMs.
- Azure AI Foundry — Microsoft. Build and run AI apps on Azure.
- Groq — Groq. Ultra-fast inference on custom LPUs.