Meta (Open Weights) · Very Fast · Updated 2026-11-22

Llama 4 Maverick API Pricing & Cost Breakdown

Open-weights portability, custom fine-tuning, and low-cost serverless or self-hosted GPU inference.

Input / 1M Tokens
$0.35
Cached Input / 1M
$0.170
Output / 1M Tokens
$1.00
Context Window
524K

Workload Cost Scenarios (30-Day Billing)

Open weights under Llama Community License; rates reflect serverless API reference.

Scenario In / Out per Req Uncached Monthly 50% Cache Hit Monthly
Hobby / Prototype (100 req/day) 1500 / 400 $2.77 $2.37
Growing Startup (2,500 req/day) 2500 / 700 $118.13 $101.25
Production Scale (50,000 req/day) 3000 / 800 $2,775.00 $2,370.00

Lower-Cost Alternative Models

GPT-6 Luna ($0.50/1M out) Claude Haiku 5.5 ($0.50/1M out) DeepSeek V4.1 Flash ($0.60/1M out)

Frequently Asked Questions

What is Llama 4 Maverick API pricing per 1M tokens?

Standard input is $0.35/1M, cached input is $0.170/1M, and output is $1.00/1M.

What is the context window of Llama 4 Maverick?

Llama 4 Maverick supports a 524K token context window with up to 32,768 max output tokens.

How much does prompt caching save on Llama 4 Maverick?

Prompt caching reduces cached input token rates by 51% compared with uncached input.

What workloads is Llama 4 Maverick best suited for?

Open-weights portability, custom fine-tuning, and low-cost serverless or self-hosted GPU inference.