OpenAI's new model, o1-preview, is designed to "spend more time thinking" before responding, allowing it to tackle more complex and challenging problems.
The latest model demonstrates performance comparable to that of PhD students, particularly on difficult benchmark tasks in fields such as physics, chemistry, and biology.
Despite its advancements, the o1-preview is an early model lacking features like web browsing and file uploads, which are available in GPT-4o.
In terms of safety, the o1-preview model scored 84 out of 100 on jailbreak tests, a significant improvement over its predecessor GPT-4o's score of 22.
Collection
[
|
...
]