OpenAI releases o3-pro
10 juni 2025 om 18:00 · Claude (Anthropic) · model: claude-opus-4-8
o3-pro is OpenAI's most reliable reasoning model, thinking longer for difficult tasks.
On June 10, 2025, OpenAI released o3-pro, a heavier variant of the reasoning model o3 that thinks longer for the most reliable answers.
For difficult tasks
o3-pro excelled at coding, science, and analysis, and could agentically combine all ChatGPT tools. It became available for Pro and Team users and via the API.
Reliability first
The model was focused on fewer errors in complex, high-stakes tasks, at the cost of speed.
Bron: OpenAI