Skip to content
llmwaves

Llama 4 Maverick

Meta · released Apr 5, 2025

Open weightsTool useStructured outputVision

Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward...

Compare this modelSource listing

Specification

Context window
1.05M
Max output
16K
Knowledge cutoff
Aug 31, 2024
Parameters
401.6B
Licence
other
Serving providers
5
Moderated
No
Uptime
100.0%

Intelligence

14.5

48th percentile

Coding

16.3

Coding Index

Agentic

1.2

Agentic Index

Output speed

72 t/s

Median across providers

Latency

324ms

Time to first token

Cost per task

$0.03

Estimated

Benchmarks

Where the score comes from

The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.

Evaluation scores

Percentage correct · higher is better

  • MMLU-Pro
    80.9%
  • GPQA Diamond
    67.1%
  • AA-LCR (long context)
    50.0%
  • IFBench
    43.0%
  • LiveCodeBench
    39.7%
  • SciCode
    33.1%
  • AIME 2025
    19.3%
  • τ²-bench (Telecom)
    17.8%
  • Terminal-Bench Hard
    6.8%
  • Humanity's Last Exam
    4.9%

An evaluation missing from this list was not run for this model — it is not a zero.

View as table
EvaluationScore
MMLU-Pro80.9%
GPQA Diamond67.1%
AA-LCR (long context)50.0%
IFBench43.0%
LiveCodeBench39.7%
SciCode33.1%
AIME 202519.3%
τ²-bench (Telecom)17.8%
Terminal-Bench Hard6.8%
Humanity's Last Exam4.9%

Against its peers

Intelligence Index · this model highlighted, nearest peers in grey

Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.

View as table
ModelIntelligence
Nemotron 3 Nano Omni (free)15.0
GPT-4.1 Mini14.8
Mistral Medium 3.114.7
Llama 4 Maverick14.5
Solar Pro 314.5
Qwen3 VL 235B A22B Instruct14.4
Ling-2.6-flash14.2
Gemini 2.5 Flash14.2

Percentile among all indexed models

Intelligence48th
Coding21th
Agentic4th
Mathematics22th
Terminal work35th
Arena Elo4th

Pricing

What it costs to run

List prices per million tokens, plus what one representative task works out to.

List price

Input / 1M tokens
$0.2
Output / 1M tokens
$0.8
Cached input / 1M
Not offered
Blended 3:1
$0.35

One task, estimated

$0.03

Input tokens
50,000
Output tokens
25,000
Profile
Standard

Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).

Serving providers

Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.

Arena

Head-to-head generation quality

Elo from pairwise judgements, broken out by the kind of thing the model was asked to build.

Overall Elo
955
Win rate
40.2%
Strongest at
3D scenes
Tournaments
164

Elo by category

Dot position on a 875–975 scale · Elo has no meaningful zero

Agent categories (full-stack apps, mobile apps) are only scored for models tested in agent harnesses.

View as table
CategoryEloWin rate
3D scenes95540.2%
UI components93440.8%
Data visualisation90938.4%
Websites89334.4%
Game development87933.7%

Same creator

More from Meta

Llama 4 Scout

Intelligence
10.3
Speed
152 t/s
Cost per task
$0.01

Llama Guard 4 12B

Intelligence
—
Speed
14 t/s
Cost per task
$0.01

Llama 3.3 70B Instruct

Intelligence
—
Speed
234 t/s
Cost per task
$0.01

Llama 3.2 1B Instruct

Intelligence
—
Speed
136 t/s
Cost per task
$0.006

Llama 3.2 3B Instruct

Intelligence
—
Speed
137 t/s
Cost per task
$0.01

Llama 3.1 70B Instruct

Intelligence
—
Speed
23 t/s
Cost per task
$0.03