Model

OpenAI o1-preview

CLOSED

OpenAI · o1 family · released Sep 12, 2024

First public o-series reasoning model, launched the think-before-answer chain-of-thought line; retired from API 2025-07-28.

ReasoningCodingVisionFunction callingTool useAgentic

1380.6

Elo · rank #141

Parameters

Undisclosed

Active params

—

Context

128K tokens

Architecture

Dense transformer reasoning model (size undisclosed)

License

Proprietary

Languages

—

API price (in/out)

$15 / $60

Modalities

text

Benchmark results

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

AIMEMath44.6%#121

best: GPT-5.2 · 100.0%

Arena-HardHuman preference81.7%#25

best: Qwen3-235B-A22B (Non-Thinking) · 96.1%

CodeforcesCoding1258#41

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

DROPReasoning74.8#38

best: Hunyuan-T1 · 93.1

GPQA DiamondReasoning73.3%#109

best: GPT-5.6 · 94.6%

HumanEvalCoding92.4%#12

best: Claude Opus 4.5 · 99.4%

MATH-500Math85.5%#57

best: GPT-5 · 99.4%

MGSMMath90.8%#8

best: OpenAI o4-mini · 93.7%

MMLUKnowledge90.8%#6

best: OpenAI o3 · 92.9%

SimpleQAKnowledge42.4%#10

best: GPT-4.5 · 62.5%

Run it locally

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

Input / M tok

$15

Output / M tok

$60

o1 family

Elo progression across releases

OpenAI o1-miniSep 20241261.7 OpenAI o1-previewSep 20241380.6 OpenAI o1Dec 20241582.8 OpenAI o1-proDec 20241700.4

Resources

DOCSdevelopers.openai.com/api/docs/models/o1-preview

API price $15/$60 · each benchmark row carries its own source badge (see methodology)