CPU compute now available
General purpose, compute optimized and memory optimized CPU instances are now available in every region, prepaid by the hour, for data preparation, retrieval, orchestration, agents and serving.
Global fabric: 13 regions
The fabric now spans 13 regions across six continents, including Middle East, Africa South and Asia South, with encrypted inter-region links and automatic failover.
Launch credit doubled
Every new account now receives $600 in launch credit plus 20M free inference tokens, double the typical big-cloud free trial.
VentureRails revenue events
Report revenue with POST /api/v1/revenue to route revenue and generate invoices automatically. Idempotent by external_id.
AI ops copilot
Ask questions about deployments, incidents and spend in plain language, with approval-gated actions and full audit history.
B200 and H200 availability
NVIDIA B200 192GB and H200 141GB instances are now available on demand with prepaid hours, right alongside our CPU instance families.
Scoped API keys
Create, scope, rotate and revoke keys from Dashboard → Security, with per-key usage analytics.