Page 25 in the paper shows user preference results. Seems it did badly vs GPT3.5 for programming topics: Paper PDF: https://www.ykilcher.com/OA_Paper_2023_04_1

Page 25 in the paper shows user preference results. Seems it did badly vs GPT3.5 for programming topics: Paper PDF: https://www.ykilcher.com/OA_Paper_2023_04_15.pdf https://www.ykilcher.com/OA_Paper_2023_04_15.pdf HN discussion on the paper: https://news.ycombinator.com/item?id=35582417 https://news.ycombinator.com/item?id=35582417 EDIT: After watching the video, seems likely this will be fine-tuned or otherwise enhanced for coding. Seems to have a lot of momentum and that's a defacto use-case.