Home/Models/Llama 4 Scout
Meta logo

Llama 4 Scout

by Meta · Released Apr 2025

Open SourceMultimodal
15.3
avg score
Rank #215
Compare
Better than 8% of all models
Context
328K tokens (~164 books)
Input $/1M
$0.08
Output $/1M
$0.30
Type
multimodal
License
Open Source
Benchmarks
11 tested
Data updated today
About

Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input...

Tested on 11 benchmarks with 18.9% average. Top scores: MATH level 5 (62.3%), Fiction.LiveBench (36.0%), GPQA diamond (35.8%).

Looking for similar performance at lower cost?
Llama 3.2 3B Instruct (free) scores 14.7 (96% as good) at $0.00/1M input · 100% cheaper
Capabilities
coding
9.1
#138 globally
reasoning
0.3
#184 globally
math
23.4
#149 globally
knowledge
35.9
#160 globally
speed
14.1
#61 globally
Benchmark Scores
Compare All
Tested on 11 benchmarks · Ranked across 5 categories
Score Distribution (all 233 models)
0255075100
▲ You are here
SWE-Bench Verified (Bash Only)

SWE-bench Verified solved using only bash commands, no specialized frameworks. Tests raw terminal-based problem solving.

9.1
ARC-AGI

Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.

0.5
ARC-AGI-2

ARC-AGI 2, harder sequel to ARC. More complex abstract reasoning patterns that test generalization ability beyond training data.

0.1
MATH level 5

Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.

62.3
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

7.7
FrontierMath-2025-02-28-Private

Original research-level math problems created by professional mathematicians. Problems are unpublished and cannot be memorized.

0.1
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Links
Documentation
Community
BenchGecko API
llama-4-scout
Specifications
  • Typemultimodal
  • Context328K tokens (~164 books)
  • ReleasedApr 2025
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.000
Available On
Meta logoMeta$0.08
Share & Export
Tweet
Llama 4 Scout is an open-source multimodal AI model by Meta, released in April 2025. It has an average benchmark score of 15.3. Context window: 328K tokens.