Meta logo
Meta · United States · Released 2026

Llama 4 Maverick

A permissive US open-weight option for full self-hosting control

Price in / 1M$0.20
Price out / 1M$0.60
Context window1M
Max output32768
LicenseOpen source

Overview

Llama 4 Maverick represents Meta’s open-weight line and remains a common default for teams that need US-based open models with broad ecosystem support. It trails closed frontier models on raw reasoning but offers strong deployment flexibility.

Best for

  • Self-hosting
  • Fine-tuning
  • US open-weight stack

Strengths & limitations

✓ Strengths

  • Permissive open-weight licence
  • Large ecosystem
  • 1M context

△ Limitations

  • Behind closed frontier on reasoning
  • Large hardware needs

Reported benchmarks

BenchmarkScoreNote
Context window~1M tokens—
LicenseOpen weights—

Benchmark figures are reported values from public sources and may use different evaluation harnesses. See sources below.

Frequently asked questions

How much does Llama 4 Maverick cost?

The API price is $0.20 per 1M input tokens and $0.60 per 1M output tokens (as of 2026-11-02). Prices change — check the official provider page for current rates.

Is Llama 4 Maverick open source?

Llama 4 Maverick is open source. Weights are available, subject to the licence terms — review the model card before commercial use.

What is the context window of Llama 4 Maverick?

Llama 4 Maverick supports a context window of 1M tokens with a maximum output of 32768 tokens per response.

What is Llama 4 Maverick best at?

Based on our scoring, Llama 4 Maverick is strongest at permissive open-weight licence, large ecosystem, 1m context. It is a common choice for self-hosting, fine-tuning, us open-weight stack.

Sources

Similar models