vLLM

About

High-throughput LLM serving engine with PagedAttention for efficient memory management.

out of 100

Data completeness46% (Minimal)

Score relies on pessimistic defaults for missing fields

Jurisdictional(25%)

Strong EU base with minimal foreign legal exposure

Ownership(20%)

Strong EU ownership and control

Bootstrapped (self-funded)
Funding: UC Berkeley research project
Bootstrapped stage
Low acquisition risk (bootstrapped, no investor pressure) (No VC exit timeline, founders retain full control)

Governance(20%)

Open, transparent governance with community involvement

Community-driven development
Backed by PyTorch Foundation
Entity control data unknown (assumed >80% single entity) (Submit verified data to improve this score)

Portability(15%)

Highly portable with self-host option and standard APIs

License(10%)

License information incomplete or mixed terms

Apache-2.0 (permissive open source)
License stability unknown (assumed at risk) (Submit verified data to improve this score)

Community(10%)

Limited community health data or concerning indicators

Bus factor unknown (assumed worst case: 1) (Submit verified data to improve this score)
Project is thriving
Release frequency unknown (assumed stale) (Submit verified data to improve this score)
Issue response time unknown (assumed >30 days) (Submit verified data to improve this score)
73K+ GitHub stars