Llama 4 Maverick vs Magistral Medium

Change models
IndexAll categories, out of 100.
34.4 ±4.7 best
33.7 ±4.1
GapIndex points below the top.
Top
−0.7 · no clear gap
CoverageIndex weight that has results.
95%
95%
SpeedOutput tokens per second.
61/s
—
EloLMArena rating and rank.
—
1255
ReleasedMonth of release.
2025-04
2025-06
Input / 1MUS dollars per 1M input tokens.
$0.35
—
Output / 1MUS dollars per 1M output tokens.
$1.00
—
1K chats2K tokens in, 500 out, each.
$1.20
—
1K documents20K tokens in, 1K out, each.
$8.00
—
1K agent tasks200K tokens in, 20K out, each.
$90.00
—
ContextMaximum tokens in one request.
524K best
40K
AgenticMulti-step tasks with tools.
36.9 best
30.6
CodingCode writing and repair.
37.2
39.2 best
ReasoningLogic problems and puzzles.
36.7 best
28.4
MultimodalTasks with images and text.
36.0
39.2 best
KnowledgeFacts and expert knowledge.
34.0
35.2 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
29.2 best
12.5
MathMath problems.
28.6
41.8 best

Benchmarks

10 shared
Only Llama 4 Maverick: 9Only Magistral Medium: 21
Llama 4 Maverick
Magistral Medium
4.9%180 / 222
9.8% best147 / 222
50.0% best154 / 217
0.0%204 / 217
0.0%150 / 214
0.3% best137 / 214
67.1%139 / 199
67.9% best138 / 199
17.8%136 / 149
23.1% best125 / 149
43.0% best94 / 146
25.1%140 / 146
0.0%120 / 131
0.0%120 / 131
4.4%123 / 126
5.9% best117 / 126
0.0%102 / 122
0.0%102 / 122
7.1%114 / 116
8.9% best109 / 116