Why OpenAI o1 Sucks at Coding [Video]

While the OpenAI’s o1 series of models is known for its exceptional reasoning capabilities, several developers have reported that these models are not the best option for programming-related tasks, especially the o1-mini.

While its slow responses continue to irk developers, the issue goes beyond response time. A developer wrote on Hacker News that the o1-preview model was hallucinating to the point where it started responding in the context of non-existing libraries and functions. 

“It’s the usual string of ‘You’re absolutely correct, and I apologise for the oversight in my previous response.’ While the reasoning may have been improved, this doesn’t solve the problem of the model having no way to assess if what it conjures up from its weights is factual or not,” he explained further.

For such reasons, developers are calling o1 models overhyped. Moutaz Alkhatib, the lead software developer at Yieldlove, mentioned that he regrets purchasing the plus tier of ChatGPT, which he bought specifically to use o1 models, …