Llama 4 Maverick

Meta Llama

· llama-4-maverick

Text and Language

Production

Lowest Cost

Open-weight workhorse with strong quality-per-dollar, deployable on any inference provider.

Best use cases: Social copy, drafts, high-volume generation

8

Quality

8

Speed

9.5

Cost Eff.

7.5

Brand Safety

7

Control

8

Adherence

Input cost

$0.22 / 1M tokens

Output cost

$0.85 / 1M tokens

Generation cost

Max context / duration

1M tokens

Aspect ratios

Resolutions

Input types

Text, images

Output types

Text

API endpoint

via OpenRouter / Together AI

Commercial use

Allowed

Data retention

Provider-dependent

Regions available

Global

API connected

Connected

Last updated

Release date

DreamCache

We make machines dream.

© 2026 DreamCache AI Ltd. All rights reserved.

The Creative Innovation OS

DreamCache AI is an independent platform. Third-party model, product and company names are trademarks of their respective owners. References to providers indicate existing or planned technical compatibility and do not imply endorsement, sponsorship or formal partnership. Model and integration availability may change according to API access, commercial terms and regional restrictions.