OpenAI o1 is the model OpenAI introduced in 2024 that pioneered inference-time long thinking. It is the release that pushed Test-time Compute into mainstream awareness — performance rose with more tokens spent per question. It is widely credited with kicking off the Reasoning Models category and was soon followed by OpenAI o3 and DeepSeek R1. On hard math, coding, and logic benchmarks it showed a clear leap over GPT-4.
MEVZU N°124ISTANBULYEAR I — VOL. III
Glossary · Beginner · 2024
OpenAI o1
OpenAI's 2024 release introducing inference-time long thinking — the first major reasoning model.
- EN — English term
- OpenAI o1
- TR — Turkish term
- OpenAI o1