how to run this model on A10 or V100

#13
by rockcat-miao - opened

this model has 10 heads, so only 2 shards will working well, but... OOM when run it on A10 and V100 with 2 shards. is this mean 10 heads dislike OLD GPUs?

Your need to confirm your account before you can post a new comment.

Sign up or log in to comment