Can i use this draft model with Q4 , Q6 and Q8 27B Models ?
#6
by hugypufy - opened
Looks like this has been asked already but can this draft model be used with Q4 , Q6 and Q8 quantised 27B models ?
If yes , does it also hold true for other models , including non-qwen ones
If no , then can you help with a guide or help point us in the direction to achieve Dflash with Q4 , Q6 and Q8