Map → L6 AI Runtime, Serving & Model Access → hosted.ai
hosted.ai
L6 · AI Runtime, Serving & Model Accessalso L5
workload abstractionstartupprivateseed
hosted.ai develops GPU virtualization and orchestration software for hosting providers and neoclouds building GPUaaS offerings, introducing multi-tenant GPU pooling, dynamic scheduling, and overcommit capabilities, plus a federated GPU Mesh marketplace for cross-operator capacity sharing.
Ecosystem functionhosted.ai is an operating system for would-be neoclouds — distinct from direct GPU lessors because it focuses on improving utilization and helping third parties launch their own GPU businesses.
Business modelSoftware licensing and capacity marketplace take-rate
Revenue—
Valuation—
Last round$19Mled by Creandum · 2026
Sources · 3 ↓
Suggest an editLast verified 2026-07-28