Provider pricing

Groq AI API pricing

Groq's LPU (Language Processing Unit) hardware delivers the fastest token throughput on the market — routinely 5–10× faster than GPU inference on the same open-weight models. GroqCloud serves Llama 4, Qwen3, DeepSeek, and Mixtral with per-token pricing competitive with the cheapest GPU providers.

Founded2016
HeadquartersMountain View, CA
Tracked models4

Best for

Real-time agents, voice assistants, and anything where time-to-first-token defines UX.

Strengths

  • 500–1000+ tokens/sec on Llama and Qwen — fastest in the industry
  • Deterministic latency from custom LPU silicon
  • OpenAI-compatible API + JSON mode + tool use
  • Free tier with generous rate limits for prototyping

Tracked Groq models

ModelInput $/MOutput $/MContextTierBest for
Llama 3.3 70B (Groq)$0.590$0.790128,000Mid-tierLPU-accelerated Llama 3.3 70B — among the fastest tokens/sec available
Llama 3.1 8B Instant (Groq)$0.050$0.080128,000BudgetUltra-fast LPU-hosted 8B for sub-second responses
Mixtral 8x7B (Groq)$0.240$0.24032,000Mid-tierLPU-hosted Mixtral 8x7B with extreme throughput
Qwen3 32B (Groq)$0.290$0.590128,000Mid-tierLPU-accelerated Qwen3 32B — extreme throughput for agents