vLLM · Pricing Plans
Vllm Plans Pricing
vLLM is free open-source software (Apache 2.0). The project does not sell hosting. Cost is incurred entirely on your own GPU infrastructure (cloud or on-prem). Several third parties (RunPod, Modal, Anyscale, Baseten, etc.) offer managed vLLM hosting, billed by them — not by the vLLM project.
1 Plans
API Commons Plans
View Source
LLMInferenceOpen SourceGPUOpenAI CompatibleSelf-HostedPlans
Plans
Self-Hosted (Apache 2.0)
free
Run vLLM on your own GPU infrastructure under Apache 2.0.
Self-Host (deployment · lifetime)
0 USD
- pip install vllm
- vllm serve
- Bring your own GPU