Inference
EN
OpenAI

GPT-4.1

GPT-4.1 is a standard model from OpenAI. 29 benchmarks count toward its index, in 8 categories.

  • Closed weights
  • Standard
  • Input: image · text · file
  • Released 2025-04
RankedEnough results for an index.
IndexAll categories, out of 100.38.0±3.0Median 51.0
CoverageIndex weight that has results.100%
SpeedOutput tokens per second.74/sMedian 67/s
First tokenSeconds to the first output token.1.0 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$2.00batch $1.00Median $0.60
Output / 1MUS dollars per 1M output tokens.$8.00batch $4.00 Batch price per 1M output tokens.Median $2.55
ContextMaximum tokens in one request.1.05MMedian 262K
EloLMArena rating and rank.1383±4 #177Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
openai/gpt-4.1
Max outputMaximum output tokens per request.
33K
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.
$0.50
WeightsLink to the published weights.
—
RetiresDate OpenRouter removes the model.
—

Where to run

APIs that serve this model.
5 providers
ProviderTypeMaker, cloud, host or router.InputOutputContext
DigitalOceanCloud$2.00$8.001.05M
Microsoft AzureDeprecatedCloud$2.00$8.001.05M
OpenRouterRouter$2.00$8.001.05M
SAP AI CoreDeprecatedCloud$2.00$8.001.05M
Snowflake CortexCloud$2.00$8.00128K

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.4 counted37.8183 / 218
CodingCode writing and repair.2 counted34.2192 / 209
ReasoningLogic problems and puzzles.5 counted40.4158 / 227
MultimodalTasks with images and text.1 counted42.1—
KnowledgeFacts and expert knowledge.6 counted41.2187 / 268
MultilingualTasks in many languages.2 counted45.16 / 9
InstructionTasks with strict prompt rules.2 counted34.248 / 63
MathMath problems.7 counted35.9135 / 168

Results

29 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
EuroEval DutchMultilingualAbove fitFar above this model's index.62.6% ±0.519 / 175
EuroEval SwedishMultilingualAbove fitFar above this model's index.63.8% ±0.723 / 195
Massive Multitask Language UnderstandingKnowledge90.2%5 / 34
EuroEval GermanMultilingual51.9% ±0.631 / 191
EuroEval FrenchMultilingualAbove fitFar above this model's index.62.3% ±0.330 / 178
EuroEval SpanishMultilingual52.8% ±0.536 / 176
EuroEval ItalianMultilingual57.1% ±0.437 / 174
EuroEval PolishMultilingual54.1% ±0.551 / 209
MATH level 5Math83.0% ±0.825 / 88
EuroEval PortugueseMultilingual54.4% ±0.554 / 182
Instruction-Following EvalInstruction87.4%34 / 70
Artificial Analysis Omniscience AccuracyKnowledge27.8%89 / 164
GPQA diamond · Epoch AIKnowledge66.9% ±2.8118 / 217
MATH 500Math87.2% ±1.533 / 59
Artificial Analysis Long Context ReasoningReasoning68.3%122 / 217
SWE-bench Verified · swebench.comCoding39.6%28 / 46
FrontierMath-2025-02-28-PrivateMath5.5% ±1.344 / 69
Terminal-Bench 1.0Agentic33.8% ±5.330 / 47
OTIS Mock AIME 2024-2025Math38.3% ±6.3128 / 200
Artificial Analysis IFBenchInstruction43.0%94 / 146
MMLU ProKnowledge80.5% ±0.487 / 135
τ²-Bench Tool-Agent-User EvaluationAgentic47.1%97 / 149
τ²-bench RetailAgentic74.0%16 / 23
MMMU ProMultimodal72.4% ±1.164 / 92
Critical Physics TasksReasoning0.0%150 / 214
Artificial Analysis GPQA DiamondKnowledge66.6%140 / 199
Graduate-Level Google-Proof Q&AKnowledge66.3%100 / 141
AIMEMath39.6% ±1.470 / 95
SimpleQA VerifiedKnowledge31.1% ±1.568 / 91
GPQA Diamond · Vals AIKnowledge65.4% ±2.4101 / 135
SWE-bench Verified · SWE-bench/experimentsCoding39.6%50 / 65
τ²-bench AirlineAgentic56.0%17 / 22
LiveCodeBenchCoding54.7% ±1.1113 / 142
Artificial Analysis MMMU-ProMultimodal61.2%90 / 113
Software Engineering Benchmark VerifiedCoding54.6%92 / 114
FrontierMath-Tier-4-2025-07-01-PrivateMath0.0%45 / 55
MMLU-ProXMultilingual72.7%14 / 17
Chess PuzzlesReasoning11.0% ±3.178 / 90
Artificial Analysis Humanity's Last ExamKnowledge4.2%194 / 222
ARC-AGI-2 (semi-private)Reasoning0.4%117 / 131
Terminal-Bench 2.0 · Vals AIAgentic14.6% ±3.862 / 66
ARC-AGI-1 (semi-private)Reasoning5.5%119 / 126
FrontierMath-Tiers-1-3-v2-PrivateMath6.0% ±1.486 / 90
SWE-Bench verified · Epoch AICoding48.5% ±2.331 / 32
τ²-bench TelecomAgentic34.0%18 / 18
Gert Labs Composite Game BenchmarkAgentic25.6%52 / 52
CaseLaw v2Legal69.9% ±0.33 / 53
FrontierMath-Tier-4-2025-07-01-PublicMath0.0%3 / 36
TaxEval v2Finance75.1% ±0.823 / 143
CorpFin v2Finance63.1% ±1.045 / 131
LegalBenchLegal83.1% ±0.458 / 149
MortgageTaxFinance65.9% ±0.938 / 97
MedQAHealthcare91.2% ±0.340 / 95
SWE-bench Multimodal · swebench.comMultimodal31.1%4 / 8
FrontierMath-2025-02-28-PublicMath0.0%33 / 64
SWE-bench Multimodal · SWE-bench/experimentsMultimodal31.1%4 / 7
MGSMMultilingual87.7% ±1.958 / 74
ARC-AGI-2 (public eval)Reasoning0.0%102 / 122
ARC-AGI-1 (public eval)Reasoning11.8%108 / 116

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseVals AINo licence statedEpoch AICC BYSierra ResearchMITEuroEvalMITSWE-bench teamNo licence statedARC Prize FoundationTerms restrict reusemodels.devMIT

History

5 releases
Index38.7 → 38.0
23 Oct 202612 Nov 2026
Output / 1M$8.00 → $8.00
23 Oct 202612 Nov 2026

Questions

How much does GPT-4.1 cost?
$2.00 per 1M input tokens, $8.00 per 1M output tokens.
How good is GPT-4.1?
Its index is 38.0 out of 100. That is place 220 of 283 ranked models.
What is GPT-4.1 best at?
Multilingual: 45.1 out of 100, place 6 of 9.
How fast is GPT-4.1?
74 output tokens per second, 1.0 s to the first token.
What is the context window of GPT-4.1?
1.05M tokens.
Is GPT-4.1 open source?
No. Its weights are not published.
When was GPT-4.1 released?
April 2025.

Badge

GPT-4.1 on Inference360: place 220 of 283, index 38.0
HTML
<a href="https://inference360.ai/models/gpt-4-1"><img src="https://inference360.ai/badges/gpt-4-1.svg" alt="GPT-4.1 on Inference360: place 220 of 283, index 38.0" height="20"></a>
Markdown
[![GPT-4.1 on Inference360: place 220 of 283, index 38.0](https://inference360.ai/badges/gpt-4-1.svg)](https://inference360.ai/models/gpt-4-1)

Compare with

North Mini Code38.0Qwen3 235B-A22B38.1DeepSeek V3.137.9GPT-4.538.8GPT-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-4.1 mini34.5GPT-4.1 nano26.3o351.8o4-mini44.7o4-mini (high)Unranked