Rentals start at $5. Pricing varies by GPU model and availability. Hourly rates are shown at checkout with budget and estimated runtime calculated for the selected configuration.
Key features
GroqMetal: dedicated bare-metal infrastructure tuned for speed and reliability
GroqCore: production-ready inference stack with no infrastructure expertise required
GroqAssured: enterprise-grade governance, auditability, and control
256 LPUs per rack with 40 PB/s SRAM bandwidth
1,000 tokens per second per user throughput
128 GB on-chip SRAM per rack
315 PFLOPS of FP8 inference compute
13 globally distributed data centers across four continents
Live GPU catalog with current inventory and pricing
Payment via x402 v2 or MPP without account signup
API, CLI, MCP server, and Agent Skill automation tools
SSH and Jupyter access to instances
Clean Ubuntu environment with NVIDIA CUDA preinstalled
Budget-based rental with transparent hourly pricing at checkout
Live ghost/stall fallback for failed provider instances
Money-back guarantee if GPU cannot be started
What makes it different
Fast inference at scale without sacrificing affordability
LPU pioneered technology integrated with NVIDIA GPUs
Fully integrated platform combining infrastructure, inference, and control
No account required for agent payments via blockchain protocols
Flexible budget-based rental with live pricing
Multiple automation interfaces for developers
Telegram-based guided rental flow
Instant provisioning after payment with credentials delivered in minutes