Llama 3.3 70B Versatile (Groq)
groq · llama-3-3-70b · official pricing
Context 128k · Max output 4k · Released 2024-12-06 · Last verified 2026-05-06
Standard pricing (per 1M tokens)
| Tier | Input | Output |
|---|---|---|
| Standard | $0.59 | $0.79 |
Capabilities
tool use caching batch extended thinking vision reasoning
Notes & extras
- Hosted by Groq. Meta provides the model weights (Llama 3.3 70B), Groq sets the hosted API price. Other hosts (Together, Fireworks, etc.) price the same model differently; check your specific host for actuals.
Sources
Found a price wrong? Report it