Comment on New OpenAI models hallucinate moreComments−8474_s1yLooks like latest chatGPT likes to hallucinate code segments that don't work or wrong API calls. Simpler instructions work fine, but at some point it starts to diverge and experiment.
Comments
Looks like latest chatGPT likes to hallucinate code segments that don't work or wrong API calls. Simpler instructions work fine, but at some point it starts to diverge and experiment.