GPT-5.2
FrontierMultimodalAPI Available
Benchmark Scores
95/100MMLU95.2
HumanEval95.8
Math93.1
Reasoning95.5
Coding95.3
About GPT-5.2
GPT-5.2 is OpenAI's latest flagship model, building on GPT-5 with improved reasoning, longer context, and enhanced agentic capabilities. It sets new records across major benchmarks and supports native tool orchestration.
Strengths
- +Best-in-class reasoning
- +512K context window
- +Native tool use and agentic workflows
- +Strongest overall benchmark performance
Weaknesses
- -Premium pricing
- -Higher latency on long contexts
- -Overkill for simple queries
Pricing
Per 1M tokens
Input$12.00
Output$36.00
Quick Facts
- Context Window
- 512K
- Parameters
- Undisclosed
- Release Date
- 2026-02-10
- Category
- Frontier
Compare Models
See how GPT-5.2 stacks up against other models