Inference.ai provides infrastructure-as-a-service GPU compute by using algorithms to match customers’ AI workloads with available GPU resources from third-party data centers. The company offers hosted GPU instances bundled with 5TB of object storage per instance. Co-founders John Yue and Michael Yu say the platform aims to simplify the confusing hardware landscape and deliver higher throughput, lower latency and lower cost compared with major public clouds. Inference claims its algorithmic matching and data-center deals enable dramatically cheaper compute and better availability, though TechCrunch did not independently verify those claims. The startup competes with providers such as CoreWeave, Lambda Labs, Together, Run.ai and Exafunction. Inference plans to use recent funding to build out its deployment infrastructure.

Total raised

$4M

Funding rounds

1

Latest round

Seed

Latest activity

Jan 2024