OpenAI-compatible API gateway priced below OpenRouter
Stage
Established
Not given
Founded
Not given
Not given
Based in
US
Not given
Pricing model
Not given
Paid
Pricing
Not given
Pay-as-you-go with no subscription. Models priced per 1M tokens in USD. For example, GPT-6 Luna: input $0.001, cache read $0.0001, output $0.005 per 1M tokens.
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
Unified API for multiple models with single base URL
OpenAI and Anthropic format compatibility
Pricing per 1M tokens with no subscription
Automatic failover between model sources
Spend limits per API key
Zero data retention for requests and responses
Two-factor authentication with backup codes
One-command setup for coding agents
Real-time model status page
Streaming support for long agent sessions
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