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
I've tried to run the inference script with a few tweaks to the python code (such as not loading the weights on the gpu twice), but I still cannot run it in bfloat16 without an OOM error. I have a RTX4070 (12GB VRAM): what are the minimum requirements for running inference with this model and are there any plan for optimizing the VRAM usage? Thanks.
The text was updated successfully, but these errors were encountered:
I've tried to run the inference script with a few tweaks to the python code (such as not loading the weights on the gpu twice), but I still cannot run it in bfloat16 without an OOM error. I have a RTX4070 (12GB VRAM): what are the minimum requirements for running inference with this model and are there any plan for optimizing the VRAM usage? Thanks.
The text was updated successfully, but these errors were encountered: