You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For those with extra VRAM, it would be nice to be able to keep some checkpoints in VRAM to be able to serve multiple models from the server without incurring the model load cost (I'm aware there's a RAM cache option but that still adds a cost)
Version Platform Description
No response
The text was updated successfully, but these errors were encountered:
Feature description
For those with extra VRAM, it would be nice to be able to keep some checkpoints in VRAM to be able to serve multiple models from the server without incurring the model load cost (I'm aware there's a RAM cache option but that still adds a cost)
Version Platform Description
No response
The text was updated successfully, but these errors were encountered: