Mistral Medium 3 vs Phi-4-reasoning
Change modelsMistral Medium 3Mistral
Phi-4-reasoningMicrosoft
IndexAll categories, out of 100.
27.1 ±5.2
27.1 ±6.4
GapIndex points below the top.
Top
Top
CoverageIndex weight that has results.
95% best
85%
SpeedOutput tokens per second.
94/s best
41/s
EloLMArena rating and rank.
1370
—
ReleasedMonth of release.
2025-05
2025-04
Input / 1MUS dollars per 1M input tokens.
$0.40
—
Output / 1MUS dollars per 1M output tokens.
$2.00
—
1K chats2K tokens in, 500 out, each.
$1.80
—
1K documents20K tokens in, 1K out, each.
$10.00
—
1K agent tasks200K tokens in, 20K out, each.
$120.00
—
ContextMaximum tokens in one request.
131K best
32K
AgenticMulti-step tasks with tools.
18.9 best
5.0
CodingCode writing and repair.
9.3
34.0 best
ReasoningLogic problems and puzzles.
32.8 best
22.1
MultimodalTasks with images and text.
24.1
—
KnowledgeFacts and expert knowledge.
31.1
33.3 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
25.7 best
21.1
MathMath problems.
32.2
35.5 best
Benchmarks
6 sharedOnly Mistral Medium 3: 8Only Phi-4-reasoning: 6
Mistral Medium 3
Phi-4-reasoning