Open vvaezian opened 1 year ago
I asked a few text-to-SQL questions from chatGPT and its answers are pretty good. I'm wondering where does it stand in the leaderboard mentioned in the readme file.
Big language models are running harder datasets like spider
For example, Codex (openai-gpt3) achieved the execution accuracy up to 67% on the Spider development set.
I asked a few text-to-SQL questions from chatGPT and its answers are pretty good. I'm wondering where does it stand in the leaderboard mentioned in the readme file.