# Llama 4 Maverick 17B Instruct API prices

Llama 4 Maverick 17B Instruct (llama) — 1.0M context, 43k max output.
8 metered per-token offers from 6 provider(s), in USD per 1M tokens.

| Provider | Input $/1M | Output $/1M | Cache read $/1M | Context |
| --- | --- | --- | --- | --- |
| Abacus | $0.14 | $0.59 | — | 1.0M |
| Charm Hyper | $0.255 | $0.8365 | $0.1275 | 430k |
| DevPass (LLM Gateway) | $0.27 | $0.85 | — | 1.0M |
| LLM Gateway | $0.27 | $0.85 | — | 1.0M |
| Amazon Bedrock | $0.24 | $0.97 | — | 1.0M |
| Amazon Bedrock (us) | $0.24 | $0.97 | — | 1.0M |
| LLM Gateway | $0.24 | $0.97 | — | 8k |
| Neon | $0.5 | $1.5 | — | 1.0M |

## Summary

- Cheapest input: $0.14 per 1M tokens (Abacus)
- Cheapest output: $0.59 per 1M tokens (Abacus)
- First-party: not listed separately
- Free offers: none
- Subscription plans (not per-token): none
- Released: 2025-04-05
- Inputs: image, text

HTML page: https://llmprice.gitlab.io/models/llama-4-maverick-17b-instruct/
Full dataset: https://llmprice.gitlab.io/data/catalog.json