out of the box probably not. gpt-3 is pretrained on a much larger corpus and is a much higher capacity model, which is why its been performing very well in zero

out of the box probably not. gpt-3 is pretrained on a much larger corpus and is a much higher capacity model, which is why its been performing very well in zero-shot learning examples like the one above. That being said you should try it. Note the huggingface library is a python library, so you'd technically be writing some code, though the library API is super user friendly and the community is very helpful.