Llama 4 Maverick vs Llama 4 Scout

Change models
IndexAll categories, out of 100.
34.4 ±4.7 best
26.8 ±4.8
GapIndex points below the top.
Top
−7.6
CoverageIndex weight that has results.
95%
95%
SpeedOutput tokens per second.
61/s
63/s best
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2025-04
2025-04
Input / 1MUS dollars per 1M input tokens.
$0.35
$0.10 best
Output / 1MUS dollars per 1M output tokens.
$1.00
$0.30 best
1K chats2K tokens in, 500 out, each.
$1.20
$0.35 best
1K documents20K tokens in, 1K out, each.
$8.00
$2.30 best
1K agent tasks200K tokens in, 20K out, each.
$90.00
$26.00 best
ContextMaximum tokens in one request.
524K
1.31M best
AgenticMulti-step tasks with tools.
36.9 best
23.6
CodingCode writing and repair.
37.2 best
23.6
ReasoningLogic problems and puzzles.
36.7 best
32.4
MultimodalTasks with images and text.
36.0 best
24.0
KnowledgeFacts and expert knowledge.
34.0 best
28.0
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
29.2 best
25.9
MathMath problems.
28.6 best
24.2

Benchmarks

18 shared
Only Llama 4 Maverick: 1
Llama 4 Maverick
Llama 4 Scout
4.9% best180 / 222
3.8%202 / 222
50.0% best154 / 217
27.7%184 / 217
67.0% best117 / 217
51.8%141 / 217
0.0%150 / 214
0.0%150 / 214
20.6% best146 / 200
7.8%160 / 200
67.1% best139 / 199
58.7%160 / 199
17.8% best136 / 149
15.5%139 / 149
43.0% best94 / 146
39.5%106 / 146
0.0%101 / 133
0.0%101 / 133
0.0%120 / 131
0.0%120 / 131
4.4% best123 / 126
0.5%125 / 126
0.0%102 / 122
0.0%102 / 122
7.1% best114 / 116
2.4%115 / 116
31.7% best103 / 114
21.3%113 / 114
62.1% best88 / 113
52.9%101 / 113
73.0% best32 / 88
62.3%41 / 88
0.7% best62 / 69
0.0%68 / 69
0.0%33 / 64
0.0%33 / 64