GPT-4.1 nano vs Phi-4-reasoning

Change models
IndexAll categories, out of 100.
26.5 ±3.9
27.2 ±6.4 best
GapIndex points below the top.
−0.7 · no clear gap
Top
CoverageIndex weight that has results.
95% best
85%
SpeedOutput tokens per second.
85/s best
34/s
EloLMArena rating and rank.
1285
—
ReleasedMonth of release.
2025-04
2025-04
Input / 1MUS dollars per 1M input tokens.
$0.10
—
Output / 1MUS dollars per 1M output tokens.
$0.40
—
Batch output / 1MBatch price per 1M output tokens.
$0.20
—
1K chats2K tokens in, 500 out, each.
$0.40
—
1K documents20K tokens in, 1K out, each.
$2.40
—
1K agent tasks200K tokens in, 20K out, each.
$28.00
—
ContextMaximum tokens in one request.
1.05M best
32K
AgenticMulti-step tasks with tools.
24.8 best
5.0
CodingCode writing and repair.
24.7
33.7 best
ReasoningLogic problems and puzzles.
31.8 best
22.4
MultimodalTasks with images and text.
17.3
—
KnowledgeFacts and expert knowledge.
27.2
33.2 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
25.3 best
21.7
MathMath problems.
29.0
35.6 best

Benchmarks

8 shared
Only GPT-4.1 nano: 27Only Phi-4-reasoning: 4
GPT-4.1 nano
Phi-4-reasoning
3.8%196 / 216
3.8%196 / 216
20.3% best187 / 211
0.0%198 / 211
0.0%145 / 207
0.0%145 / 207
51.2%166 / 193
57.5% best157 / 193
50.3%127 / 144
65.8% best100 / 144
17.3% best131 / 143
0.0%142 / 143
32.0% best123 / 140
23.5%136 / 140
83.2%45 / 69
83.4% best42 / 69