Inference
EN
OpenAI

GPT-4.1 nano

GPT-4.1 nano is a standard model from OpenAI in the GPT-4.1 family. 22 benchmarks count toward its index, in 7 categories.

  • Closed weights
  • Standard
  • Input: image · text · file
  • Released 2025-04
RankedEnough results for an index.
IndexAll categories, out of 100.26.3±3.9Median 51.0
CoverageIndex weight that has results.95%
SpeedOutput tokens per second.56/sMedian 67/s
First tokenSeconds to the first output token.1.4 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$0.10batch $0.05Median $0.60
Output / 1MUS dollars per 1M output tokens.$0.40batch $0.20 Batch price per 1M output tokens.Median $2.55
ContextMaximum tokens in one request.1.05MMedian 262K
EloLMArena rating and rank.1285±8 #269Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
openai/gpt-4.1-nano
Max outputMaximum output tokens per request.
943K
Knowledge cutoffLast date of training data.
30 Jun 2024
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
Yes
Cache read / 1MUS dollars per 1M cached tokens.
$0.03
WeightsLink to the published weights.
—
RetiresDate OpenRouter removes the model.
—

Where to run

APIs that serve this model.
3 providers
ProviderTypeMaker, cloud, host or router.InputOutputContext
Microsoft AzureDeprecatedCloud$0.10$0.401.05M
OpenRouterRouter$0.10$0.401.05M
SAP AI CoreDeprecatedCloud$0.10$0.401.05M

US dollars per 1M tokens, list price. From models.dev (opens in a new tab) (MIT).

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.2 counted24.2216 / 218
CodingCode writing and repair.1 counted24.8—
ReasoningLogic problems and puzzles.5 counted31.6202 / 227
MultimodalTasks with images and text.1 counted17.1—
KnowledgeFacts and expert knowledge.6 counted26.7251 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.2 counted24.859 / 63
MathMath problems.5 counted29.1145 / 168

Results

22 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MATH level 5Math70.0% ±1.033 / 88
Instruction-Following EvalInstruction83.2%46 / 70
MATH 500Math80.2% ±1.839 / 59
OTIS Mock AIME 2024-2025Math28.9% ±6.0136 / 200
GPQA diamond · Epoch AIKnowledge48.9% ±2.5149 / 217
Critical Physics TasksReasoning0.0%150 / 214
GDPval-AA normalizedAgentic0.0%101 / 133
Massive Multitask Language UnderstandingKnowledge80.1%26 / 34
AIMEMath26.5% ±0.876 / 95
FrontierMath-2025-02-28-PrivateMath1.0% ±0.659 / 69
Artificial Analysis GPQA DiamondKnowledge51.2%172 / 199
GPQA Diamond · Vals AIKnowledge50.8% ±2.5118 / 135
Graduate-Level Google-Proof Q&AKnowledge50.3%124 / 141
LiveCodeBenchCoding42.7% ±1.2125 / 142
Artificial Analysis IFBenchInstruction32.0%129 / 146
Artificial Analysis Long Context ReasoningReasoning20.3%193 / 217
Artificial Analysis Humanity's Last ExamKnowledge3.8%202 / 222
ARC-AGI-2 (semi-private)Reasoning0.0%120 / 131
τ²-Bench Tool-Agent-User EvaluationAgentic17.3%137 / 149
Artificial Analysis Omniscience AccuracyKnowledge13.7%156 / 164
MMMU ProMultimodal55.1% ±1.288 / 92
MMLU ProKnowledge63.5% ±0.5130 / 135
Artificial Analysis MMMU-ProMultimodal40.1%110 / 113
SimpleQA VerifiedKnowledge6.0% ±0.891 / 91
ARC-AGI-1 (semi-private)Reasoning0.0%126 / 126
Chess PuzzlesReasoning0.0%90 / 90
FrontierMath-2025-02-28-PublicMath0.0%33 / 64
MortgageTaxFinance52.8% ±1.081 / 97
ARC-AGI-2 (public eval)Reasoning0.0%102 / 122
MedQAHealthcare68.2% ±0.483 / 95
TaxEval v2Finance60.8% ±1.0126 / 143
CorpFin v2Finance42.1% ±1.0121 / 131
LegalBenchLegal61.1% ±0.6141 / 149
MGSMMultilingual69.3% ±2.873 / 74
ARC-AGI-1 (public eval)Reasoning1.8%116 / 116

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseVals AINo licence statedEpoch AICC BYARC Prize FoundationTerms restrict reusemodels.devMIT

History

5 releases
Index27.1 → 26.3
23 Oct 202612 Nov 2026
Output / 1M$0.40 → $0.40
23 Oct 202612 Nov 2026

Questions

How much does GPT-4.1 nano cost?
$0.10 per 1M input tokens, $0.40 per 1M output tokens.
How good is GPT-4.1 nano?
Its index is 26.3 out of 100. That is place 265 of 283 ranked models.
What is GPT-4.1 nano best at?
Math: 29.1 out of 100, place 145 of 168.
How fast is GPT-4.1 nano?
56 output tokens per second, 1.4 s to the first token.
What is the context window of GPT-4.1 nano?
1.05M tokens.
Is GPT-4.1 nano open source?
No. Its weights are not published.
When was GPT-4.1 nano released?
April 2025.

Badge

GPT-4.1 nano on Inference360: place 265 of 283, index 26.3
HTML
<a href="https://inference360.ai/models/gpt-4-1-nano"><img src="https://inference360.ai/badges/gpt-4-1-nano.svg" alt="GPT-4.1 nano on Inference360: place 265 of 283, index 26.3" height="20"></a>
Markdown
[![GPT-4.1 nano on Inference360: place 265 of 283, index 26.3](https://inference360.ai/badges/gpt-4-1-nano.svg)](https://inference360.ai/models/gpt-4-1-nano)

Compare with

GPT-4o mini26.2Llama 4 Scout26.8Nova Pro25.6Llama 3.1 405B27.0GPT-4.1 mini34.5

Lower price

dots3-note Preview63.4 · FreeApodex 1.1 Mini55.7 · FreeNemotron 3 Ultra50.8 · Free

More from OpenAI

All models
GPT-4.138.0GPT-4.1 mini34.5o351.8o4-mini44.7o4-mini (high)Unranked