Llama 4 Maverick

Creator

Meta AI (FAIR)

Released

2025-04-05

Intelligence

14.3

Artificial Analysis Index

Coding

16.3

Artificial Analysis Index

In $/1M

$0.27

input tokens

Out $/1M

$0.85

output tokens

Blended $/1M

$0.41

3:1 blended

Speed

106

tokens / sec

Profile

License, openness, and modality — the governance layer.

License

Llama 4 Community License Agreement

non-commercial

Weights

Open

downloadable

Modalities

Image · Text

Parameters

400B

total

Llama 4 Maverick is a natively multimodal model capable of processing both text and images. It features a 17 billion active parameter mixture-of-experts (MoE) architecture with 128 experts, supporting a wide range of multimodal tasks such as conversational interaction, image analysis, and code generation. The model includes a 1 million token context window.

Capability profile

Category strength across 20 domains, via LLM Stats.

image to text
90math
80legal
80vision
80finance
80general
80language
80
healthcare
80multimodal
80biology
70physics
70chemistry
70reasoning
70code
40
Via LLM Stats

Benchmark breakdown

Independent evaluation scores, normalized to 0–100.

GPQA Diamond
67
Humanity’s Last Exam
5
MMLU-Pro
SciCode
33
LiveCodeBench
40
MATH-500
89
AIME 2025
19
τ²-Bench (agentic)
18
Terminal-Bench Hard
7
IFBench
43
Via Artificial Analysis
© 2026 NYSGPT2525 LLC