Inference
EN
OpenAI

GPT-4o

GPT-4o is a standard model from OpenAI. 24 benchmarks count toward its index, in 8 categories.

  • Closed weights
  • Standard
  • Input: text · image · file
  • Released 2024-05
RankedEnough results for an index.
IndexAll categories, out of 100.28.7±3.3Median 51.0
CoverageIndex weight that has results.100%
SpeedOutput tokens per second.66/sMedian 67/s
First tokenSeconds to the first output token.0.6 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$2.50batch $1.25Median $0.60
Output / 1MUS dollars per 1M output tokens.$10.00batch $5.00 Batch price per 1M output tokens.Median $2.55
ContextMaximum tokens in one request.128KMedian 262K
EloLMArena rating and rank.1300±3 #252Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
openai/gpt-4o
Max outputMaximum output tokens per request.
16K
Knowledge cutoffLast date of training data.
31 Oct 2023
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
Yes
Cache read / 1MUS dollars per 1M cached tokens.
$1.25
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
DigitalOceanCloud$2.50$10.00128K
Microsoft AzureDeprecatedCloud$2.50$10.00128K
OpenRouterRouter$2.50$10.00128K

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.2 counted26.7208 / 218
CodingCode writing and repair.3 counted22.8205 / 209
ReasoningLogic problems and puzzles.5 counted37.4178 / 227
MultimodalTasks with images and text.1 counted36.0—
KnowledgeFacts and expert knowledge.5 counted34.1229 / 268
MultilingualTasks in many languages.1 counted30.8—
InstructionTasks with strict prompt rules.1 counted21.1—
MathMath problems.6 counted23.4155 / 168

Results

24 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
SWE-bench Lite · SWE-bench/experimentsCoding28.0%8 / 15
MATH level 5Math51.0% ±1.148 / 88
SWE-bench Lite · swebench.comCoding21.7%9 / 15
GPQA diamond · Epoch AIKnowledge48.9% ±2.6149 / 217
Critical Physics TasksReasoning0.0%150 / 214
Artificial Analysis Long Context ReasoningReasoning49.3%158 / 217
GDPval-AA normalizedAgentic0.0%101 / 133
MATH 500Math74.0% ±3.846 / 59
Artificial Analysis Omniscience AccuracyKnowledge19.9%131 / 164
SimpleQA VerifiedKnowledge26.0% ±1.473 / 91
SWE-bench Verified · swebench.comCoding23.2%37 / 46
τ²-Bench Tool-Agent-User EvaluationAgentic25.1%120 / 149
OTIS Mock AIME 2024-2025Math6.3% ±2.3166 / 200
MMLU ProKnowledge72.6% ±0.4114 / 135
Artificial Analysis GPQA DiamondKnowledge54.3%169 / 199
Artificial Analysis IFBenchInstruction34.3%125 / 146
LiveCodeBenchCoding43.4% ±1.1123 / 142
GPQA Diamond · Vals AIKnowledge53.8% ±2.5117 / 135
MMMU ProMultimodal62.2% ±1.281 / 92
Chess PuzzlesReasoning9.0% ±2.981 / 90
AIMEMath11.9% ±1.086 / 95
SWE-bench Verified · SWE-bench/experimentsCoding21.6%59 / 65
ARC-AGI-2 (semi-private)Reasoning0.0%120 / 131
FrontierMath-2025-02-28-PrivateMath0.3% ±0.364 / 69
MMLU-ProXMultilingual61.1%16 / 17
ARC-AGI-1 (semi-private)Reasoning4.5%122 / 126
FrontierMath-Tiers-1-3-v2-PrivateMath0.4% ±0.489 / 90
SWE-Bench verified · Epoch AICoding31.0% ±2.132 / 32
Artificial Analysis Humanity's Last ExamKnowledge2.4%222 / 222
TaxEval v2Finance74.5% ±0.935 / 143
CaseLaw v2Legal59.7% ±0.425 / 53
LegalBenchLegal82.2% ±0.574 / 149
MGSMMultilingual90.4% ±1.838 / 74
FrontierMath-2025-02-28-PublicMath0.0%33 / 64
SWE-bench (full test split)Coding14.6%4 / 7
MedQAHealthcare88.2% ±0.655 / 95
MortgageTaxFinance57.4% ±1.975 / 97
ARC-AGI-2 (public eval)Reasoning0.0%102 / 122
SWE-bench Multimodal · swebench.comMultimodal10.6%7 / 8
CorpFin v2Finance45.9% ±1.0115 / 131
Finance Agent v1.1Finance8.1% ±1.547 / 51
SWE-bench Multimodal · SWE-bench/experimentsMultimodal6.0%7 / 7

Sources

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

History

5 releases
Index29.4 → 28.7
23 Oct 202612 Nov 2026
Output / 1M$10.00 → $10.00
23 Oct 202612 Nov 2026

Questions

How much does GPT-4o cost?
$2.50 per 1M input tokens, $10.00 per 1M output tokens.
How good is GPT-4o?
Its index is 28.7 out of 100. That is place 254 of 283 ranked models.
What is GPT-4o best at?
Reasoning: 37.4 out of 100, place 178 of 227.
How fast is GPT-4o?
66 output tokens per second, 0.6 s to the first token.
What is the context window of GPT-4o?
128K tokens.
Is GPT-4o open source?
No. Its weights are not published.
When was GPT-4o released?
May 2024.

Badge

GPT-4o on Inference360: place 254 of 283, index 28.7
HTML
<a href="https://inference360.ai/models/gpt-4o"><img src="https://inference360.ai/badges/gpt-4o.svg" alt="GPT-4o on Inference360: place 254 of 283, index 28.7" height="20"></a>
Markdown
[![GPT-4o on Inference360: place 254 of 283, index 28.7](https://inference360.ai/badges/gpt-4o.svg)](https://inference360.ai/models/gpt-4o)

Compare with

Grok 229.1Llama 3.3 70B28.2Devstral Small28.2LFM2.5-2.6B29.3GPT-4o mini26.2

Lower price

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

More from OpenAI

All models
GPT-4o mini26.2GPT-3.5 Turbo (older v0613)Unrankedo1-mini33.5o1-previewUnrankedGPT-4o AudioUnranked