Benchmark Scores
93.2/100MMLU91.8
HumanEval92.4
Math94.8
Reasoning95.1
Coding92
About OpenAI o1
OpenAI o1 is a reasoning-focused model that uses chain-of-thought to solve complex problems. It excels at math, science, and coding competitions, achieving PhD-level performance on many academic benchmarks.
Strengths
- +Exceptional mathematical reasoning
- +Strong at complex multi-step problems
- +PhD-level science accuracy
- +Excellent competition math
Weaknesses
- -High latency due to chain-of-thought
- -Expensive
- -Not ideal for simple conversational tasks
Pricing
Per 1M tokens
Input$15.00
Output$60.00
Quick Facts
- Context Window
- 200K
- Parameters
- Undisclosed
- Release Date
- 2024-12-17
- Category
- Frontier
Compare Models
See how OpenAI o1 stacks up against other models