Model

OpenAI o3-mini (Medium)

CLOSED

OpenAI · o3 family · released Jan 31, 2025

Default reasoning-effort setting; roughly matches full o1 on AIME/GPQA at a fraction of the cost.

ReasoningCodingVisionFunction callingTool useAgentic

1485.5

Elo · rank #123

Parameters

Undisclosed

Active params

—

Context

200K tokens

Architecture

Dense transformer reasoning model (size undisclosed)

License

Proprietary

Languages

—

API price (in/out)

$1.1 / $4.4

Modalities

text

Benchmark results

Bar shows position within the tracked field; marker = field best

AIMEMath76.9%#87

best: GPT-5.2 · 100.0%

CodeforcesCoding2036#18

best: DeepSeek-V4-Pro (Think Max) · 3206

DROPReasoning79.2#29

best: Hunyuan-T1 · 93.1

GPQA DiamondReasoning76.8%#94

best: GPT-5.6 · 94.6%

HumanEvalCoding96.3%#4

best: Claude Opus 4.5 · 99.4%

MATH-500Math97.3%#18

best: GPT-5 · 99.4%

MGSMMath90.8%#9

best: OpenAI o4-mini · 93.7%

MMLUKnowledge85.9%#51

best: OpenAI o3 · 92.9%

PutnamBenchMath0.0%#11

best: DeepSeek-Prover-V2 671B · 7.1%

SimpleQAKnowledge13.4%#29

best: GPT-4.5 · 62.5%

SWE-bench VerifiedCoding42.9%#98

best: Claude Fable 5 · 95.0%

Run it locally

Closed weights — available via API only. No local deployment.

Input / M tok

$1.1

Output / M tok

$4.4

o3 family

Elo progression across releases

OpenAI o3-mini (High)Jan 20251653.0 OpenAI o3-mini (Low)Jan 20251334.3 OpenAI o3-mini (Medium)Jan 20251485.5 OpenAI o3Apr 20251926.8 OpenAI o3-proJun 20252103.6

Resources

DOCSdevelopers.openai.com/api/docs/models/o3-mini APIAvailable via OpenAI API

API price $1.1/$4.4 · each benchmark row carries its own source badge (see methodology)