OpenAIFeatured

GPT-5.2 Thinking

OpenAI's advanced reasoning model. Excels at complex structured work like coding, analyzing long documents, math, and planning. Beats or ties top industry professionals on 70.9% of GDPval knowledge work tasks.

4.9/ 5.0
128K context
Try GPT-5.2 o1

Pricing

$200/month
Pro plan
API Pricing
Input$15/1M tokens
Output$60/1M tokens

Ratings

Overall
4.9
Coding
4.8
Writing
4.5
Reasoning
5.0
Speed
3.8
Value
4.0

Pros

  • Best reasoning model from OpenAI
  • Edges out Gemini 3 and Claude Opus 4.5 on most tests
  • Performs at/above human expert level on GDPval
  • Excellent for coding, math, and document analysis
  • State-of-the-art on SWE-Bench Pro

Cons

  • Expensive ($200/mo for Pro subscription)
  • Slower than Instant variant
  • Requires Pro subscription for full access
  • Writing can feel robotic

Benchmark Results

SWE-bench Verified
76.3%#3
LMArena Elo
1478Elo#3
AIME 2025
94.6%#2
GDPval
70.9%#1

Best For

Complex reasoning, software engineering, research, math proofs

Quick Info

Provider
OpenAI
Version
5.2
Released
Dec 2025
Context
128K tokens

Features

Vision/Images
Audio
Video
PDF Processing
Code Generation
AI Agents

Ready to try GPT-5.2 o1?

Get started with a pro account.

Get Started