> Llama 4 Maverick is an open-weights AI model from Meta, released 5 Apr 2025. Index score 115 (GPT-4 = 100), rank 149 of 229 ranked models. API price $0.19 in / $0.65 out per million tokens, 1.05M context.

[Models](https://themodelindex.org/models/) /[Meta](https://themodelindex.org/labs/meta/)

# Llama 4 Maverick

Meta

Released 5 Apr 2025

Open

Date verified

[Compare](https://themodelindex.org/compare/llama-4-maverick-vs-o3-2025-04/) [Announcement](https://about.fb.com/ltam/news/2025/04/la-coleccion-de-modelos-llama-4-el-inicio-de-una-nueva-era-de-innovacion-multimodal-nativa-para-inteligencia-artificial/)

115104–126

Index score

#149 of 229 models

3/ 8

Core tests taken

2 of 6 areas · plus 12 anchor tests

−25

Versus the frontier at release

best then: Gemini 2.5 Pro

$0.19/ $0.65

Price per million tokens

input / output, list price

1.05M

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

o1-mini reached this score 7 months earlier.

## Test results · by area · tick = best by any model

Show 10 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

67%

Humanity's Last Exam

6%

Maths

evidence

FrontierMath T1-3

no result

Coding

evidence

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

evidence

APEX-Agents

no result

Novel problems

evidence

ARC-AGI-2

0%

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#149

of 219 · 115

Artificial Analysis Intelligence Index

overall capability across ten evaluations

#111

of 207 · 10

Epoch Capabilities Index

overall capability from many benchmarks

#129

of 239 · 132.2

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

Not in the top three among current models on any category we track.

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

Meta AI

app · subscription

↗

#### API providers · 5 · per million tokens, in / out

D

DigitalOcean

$0.19 / $0.65

cheapest

D

DeepInfra

$0.20 / $0.80 · fp8

↗

N

Novita

$0.27 / $0.85 · fp8

↗

P

Parasail

$0.35 / $1.0 · fp8

↗

G

Google

$0.35 / $1.15

↗

#### Run it yourself

🤗

Official weights

meta-llama/Llama-4-Maverick-17B-128E-Instruct

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

5 Apr 2025

Announcement

[about.fb.com ↗](https://about.fb.com/ltam/news/2025/04/la-coleccion-de-modelos-llama-4-el-inicio-de-una-nueva-era-de-innovacion-multimodal-nativa-para-inteligencia-artificial/)

Release entry

Llama 4 (Scout/Maverick)

Scored as

Llama 4 Maverick

Artificial Analysis index

10 (their estimate)

Output speed

84 tokens/s

Hugging Face

[meta-llama/Llama-4-Maverick-17B-128E-Instruct ↗](https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E-Instruct)

## Meta releases · around this one

Llama 3.2

Open

25 SEP 2024

index 103

Llama 3.3 70B

Open

6 DEC 2024

index 106

Llama 4 Maverick

Open

5 APR 2025

index 115

Llama 4 Scout

Open

5 APR 2025

index 111

Muse Spark

Closed

8 APR 2026

index 159

---
Source: https://themodelindex.org/models/llama-4-maverick/ · The Model Index · data as of 2026-10-01