Llama 4 Scout
#160 of 274 scored.
MetaApr 5, 2025Open weights
Capabilities Index
Range 124.3–131.4
129.6
Context
Takes text, image
1.31M
Input
Per 1M tokens
$0.10
Output
Per 1M tokens
$0.30
Benchmarks
How it scores.
- GPQA Diamond35.8%
130th of 186
- ARC-AGI-20.0%
72nd of 79
- Mock AIME7.7%
139th of 176
Everything else Epoch has for it
- MATH level 562.3%
- Fiction.LiveBench36.0%
- DTBench29.8%
- LMCA14.1%
- ARC-AGI0.5%
- FrontierMath-2025-02-28-Private0.0%
On the index
Around it.
Sources