OpenAI has introduced the o3-mini model, enhancing developer capabilities through features like function calling and structured outputs. Developers can also select a reasoning effort level to customize performance. The model, however, lacks vision capabilities compared to its predecessor. OpenAI emphasizes o3-mini's safety, asserting it outperforms previous versions in safety evaluations. The model is currently available in various APIs, focusing on developers positioned in higher usage tiers, ensuring robust assistance in software integration.
The model also offers new features for developers who incorporate OpenAI models in their software, including function calling, developer messages, and structured outputs.
Before deployment, we carefully assessed the safety risks of o3-mini using the same approach to preparedness, external red-teaming, and safety evaluations as o1.
Collection
[
|
...
]