Groq API pricing
6 Groq models, in US dollars per million tokens. Updated Oct 29, 2026.
6 models
| Model | Input | Cached input | Output | Context | Notes |
|---|---|---|---|---|---|
| meta-llama/llama-prompt-guard-2-22m | $0.03 | — | $0.03 | 1K | |
| meta-llama/llama-prompt-guard-2-86m | $0.04 | — | $0.04 | 1K | |
| openai/gpt-oss-120b | $0.15 | $0.075 | $0.60 | 131K | |
| openai/gpt-oss-20b | $0.075 | $0.0375 | $0.30 | 131K | |
| openai/gpt-oss-safeguard-20b | $0.075 | $0.037 | $0.30 | 131K | |
| qwen/qwen3.8-27b | $0.80 | — | $4.00 | 131K |
See what your Groq calls actually cost, by service, model and customer, priced for where each call ran.
Get started freeList prices from LiteLLM's public price file. Long-prompt, batch and priority rates aren't shown; check Groq's pricing page before you commit to a budget. All providers