Last week there was a post about Microsoft's FastContext paper: https://www.reddit.com/r/LocalLLaMA/comments/1ud1lro/why_is_no_one_talking_about_microsofts_open/ I commented that I wanted to run a related benchmark with my own setup. This is the follow-up. This is not a direct comparison against Fas
