Comment by segmondy 1 year ago how much context size? 1 comment segmondy Reply smcleod 1 year ago Just 4K. Because deepseek doesn't allow for the use of flash attention it means you can't run quantised qkv
smcleod 1 year ago Just 4K. Because deepseek doesn't allow for the use of flash attention it means you can't run quantised qkv
Just 4K. Because deepseek doesn't allow for the use of flash attention it means you can't run quantised qkv