2 ms·
Their smallest model outperforms GPT-4 on Code. I'm sceptical that it'll hold up to real world use though.
by hackerlight 3y ago
Their smallest model outperforms GPT-4 on Code. I'm sceptical that it'll hold up to real world use though.
- nopinsight 3y agoJust a note that the 67.0% HumanEval figure for GPT-4 is from its first release in March 2023. The actual performance of current ChatGPT-4 on similar problems might be better due to OpenAI's internal system prompts, possible fine-tuning, and other tricks.