We are currently using llama3-8b. We might want to try using 3.1.
We are currently using llama3-8b. We might want to try using 3.1.