Skip to content
AI Primer

o3

Reasoning model for complex tasks, succeeded by GPT-5

o3 is OpenAI's reasoning model release for complex tasks across text, code, and images. OpenAI's API documentation describes it as well-rounded and powerful across domains, with a 200,000-token context window, text and image input, text output, and strengths in math, science, coding, technical writing, instruction following, and visual reasoning.

Pricing

Model profile · Current snapshot
Input / 1M
$2.00
Output / 1M
$8.00
Blended / 1M
$3.50
Output TPS
128
TTFT (s)
5.51

Model Intelligence

Context window
200,000 tokens
Arena ranking
31
Benchmarkable
Yes
Model level
release
Intelligence Index
31.1
Math Index
88.3
MMLU Pro
0.85
GPQA
0.83
HLE
0.2
LiveCodeBench
0.81
SciCode
0.41
MATH-500
0.99
AIME
0.9
AIME 2025
0.88
IFBench
0.71
LCR
0.73
TerminalBench Hard
0.37
TAU2
0.81

Recent stories

0 linked stories
No linked stories yet.
AI PrimerAI Primer

Your daily guide to AI tools, workflows, and creative inspiration.

© 2026 AI Primer. All rights reserved.