Inference
EN
OpenAI

o3-pro

o3-pro is a reasoning model from OpenAI in the o3 family. It has 3 scored results in 2 categories. It has no index.

  • Closed weights
  • Reasoning
  • Input: text · file · image
  • Released 2025-06
  • Retires 11 Dec 2026
UnrankedEnough results for an index.
IndexAll categories, out of 100.UnrankedMedian 51.0
CoverageIndex weight that has results.30%
SpeedOutput tokens per second.15/sMedian 67/s
First tokenSeconds to the first output token.5.7 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$20.00Median $0.60
Output / 1MUS dollars per 1M output tokens.$80.00Median $2.55
ContextMaximum tokens in one request.200KMedian 262K
EloLMArena rating and rank.—Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
openai/o3-pro
Max outputMaximum output tokens per request.
100K
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.
—
WeightsLink to the published weights.
—
RetiresDate OpenRouter removes the model.
11 Dec 2026

Where to run

APIs that serve this model.
1 provider
ProviderTypeMaker, cloud, host or router.InputOutputContext
OpenRouterRouter$20.00$80.00200K

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.None counted——
CodingCode writing and repair.None counted——
ReasoningLogic problems and puzzles.2 counted47.8—
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.1 counted54.7—
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.None counted——

Results

3 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Artificial Analysis GPQA DiamondKnowledge84.5%80 / 199
ARC-AGI-1 (semi-private)Reasoning59.3%68 / 126
ARC-AGI-2 (semi-private)Reasoning4.9%80 / 131
ARC-AGI-1 (public eval)Reasoning63.3%65 / 116
ARC-AGI-2 (public eval)Reasoning3.9%76 / 122

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseARC Prize FoundationTerms restrict reusemodels.devMIT

History

5 releases
Output / 1M$80.00 → $80.00
23 Oct 202612 Nov 2026

Questions

How much does o3-pro cost?
$20.00 per 1M input tokens, $80.00 per 1M output tokens.
How fast is o3-pro?
15 output tokens per second, 5.7 s to the first token.
What is the context window of o3-pro?
200K tokens.
Is o3-pro open source?
No. Its weights are not published.
When was o3-pro released?
June 2025.

More from OpenAI

All models
o351.8o4-mini44.7o4-mini (high)UnrankedGPT-OSS 120B43.4GPT-OSS 20B41.7