undefined | Better HN

0 pointsjmiskovic2y ago0 comments

Then why not read their paper?

"The largest Llama 2-Chat model is competitive with ChatGPT. Llama 2-Chat 70B model has a win rate of 36% and a tie rate of 31.5% relative to ChatGPT."

0 pointsjmiskovic2y ago0 comments

Then why not read their paper?

"The largest Llama 2-Chat model is competitive with ChatGPT. Llama 2-Chat 70B model has a win rate of 36% and a tie rate of 31.5% relative to ChatGPT."