# Llama 4 Maverick 17B 128E Instruct FP8 API prices

Llama 4 Maverick 17B 128E Instruct FP8 (llama) — 1.0M context, 16k max output.
3 metered per-token offers from 5 provider(s), in USD per 1M tokens.

| Provider | Input $/1M | Output $/1M | Cache read $/1M | Context |
| --- | --- | --- | --- | --- |
| Azure | $0.25 | $1 | — | 1.0M |
| Azure Cognitive Services | $0.25 | $1 | — | 1.0M |
| watsonx.ai | $0.371 | $1.484 | — | 131k |

## Summary

- Cheapest input: $0.25 per 1M tokens (Azure)
- Cheapest output: $1 per 1M tokens (Azure)
- First-party: not listed separately
- Free offers: Llama, Vercel AI Gateway
- Subscription plans (not per-token): none
- Released: 2025-04-05
- Inputs: image, text

HTML page: https://llmprice.gitlab.io/models/llama-4-maverick-17b-128e-instruct-fp8/
Full dataset: https://llmprice.gitlab.io/data/catalog.json