Llama 4 Maverick

Available

Meta · 2025-04-05 · 32,000 tokens

An AI model from Meta, suited to a broad range of AI workloads.

Supported modalities:textcode

Quick Overview

Text Generation1/10

Code Generation2/10

Reasoning2/10

Multimodal1/10

Scores from leading benchmark suites.

artificial analysis intelligence14.5

artificial analysis coding16.3

artificial analysis math19.3

Latency and throughput performance.

—P50 Latency

110.267tokens/sec