GPT-5.1 vs GPT-5 nano

Change models
GPT-5.1OpenAI
IndexAll categories, out of 100.
54.6 ±3.7 best
41.7 ±6.2
GapIndex points below the top.
Top
−12.9
CoverageIndex weight that has results.
95% best
75%
SpeedOutput tokens per second.
66/s
147/s best
EloLMArena rating and rank.
1422 best
1320
ReleasedMonth of release.
2025-11
2025-08
Input / 1MUS dollars per 1M input tokens.
$1.25
$0.05 best
Output / 1MUS dollars per 1M output tokens.
$10.00
$0.40 best
Batch output / 1MBatch price per 1M output tokens.
$5.00
$0.20 best
1K chats2K tokens in, 500 out, each.
$7.50
$0.30 best
1K documents20K tokens in, 1K out, each.
$35.00
$1.40 best
1K agent tasks200K tokens in, 20K out, each.
$450.00
$18.00 best
ContextMaximum tokens in one request.
400K
400K
AgenticMulti-step tasks with tools.
52.0
—
CodingCode writing and repair.
52.9 best
37.1
ReasoningLogic problems and puzzles.
55.4 best
42.5
MultimodalTasks with images and text.
57.8 best
47.1
KnowledgeFacts and expert knowledge.
56.7 best
37.4
MultilingualTasks in many languages.
—
44.5
InstructionTasks with strict prompt rules.
56.3
—
MathMath problems.
56.8 best
45.4

Benchmarks

21 shared
Only GPT-5.1: 23Only GPT-5 nano: 19
GPT-5.1
GPT-5 nano
87.6% best65 / 220
69.4%117 / 220
88.6% best70 / 203
81.1%88 / 203
85.7% best20 / 149
50.1%147 / 149
TaxEval v2Finance
74.9% best26 / 143
67.4%107 / 143
86.5% best22 / 142
70.2%88 / 142
86.6% best41 / 135
63.4%105 / 135
MMLU ProKnowledge
86.4% best42 / 135
76.1%107 / 135
MedScribeHealthcare
88.1% best15 / 111
72.9%87 / 111
MedCodeHealthcare
52.7% best13 / 108
30.4%96 / 108
61.4% best60 / 97
53.6%80 / 97
AIMEMath
93.3% best15 / 95
81.2%45 / 95
MedQAHealthcare
96.4% best2 / 95
93.3%23 / 95
48.0% best38 / 94
11.7%90 / 94
Chess PuzzlesReasoning
42.0% best36 / 93
22.0%60 / 93
MMMU ProMultimodal
83.2% best33 / 92
70.9%69 / 92
16.0% best59 / 83
8.0%69 / 83
MGSMMultilingual
93.0% best11 / 74
89.3%47 / 74
31.0% best18 / 69
8.3%40 / 69
66.0% best19 / 65
34.8%53 / 65
12.5% best17 / 55
0.0%45 / 55