Search for a command to run...
Explore token-based, subscription, credit, and compute pricing for Groq. Data snapshot: 2025-10-17.
Token • per 1M tokens
Input $0.050 · Output $0.080
Ultra-low latency tier
Latency guarantees typically <200ms for 8B; pricing captured 2024-09.
View provider catalogToken • per 1M tokens
Input $0.590 · Output $0.790
High-accuracy, hardware-accelerated tier
Compute • $0.000 per second
Effective per-second compute with 500 tok/s target
Pricing captured 2024-09; compute estimate derived from Groq docs.
View provider catalogToken • per 1M tokens
Input $0.270 · Output $0.400
Mixture-of-experts with deterministic throughput
Pricing captured 2024-09.
View provider catalogCurated data is maintained manually; upstream plan changes may require confirmation before automation catches them.