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
Currently, we know that the DLRM cuda libraries written by NVIDIA can run on V100 (compute capability 7.0) and A100 GPUs (compute capability 8.0). If we move to a newer GPU, such as RTX 3090 (compute capability 8.6), would the existing code work there? We might also need to change the nvcc flags for compilation.
The text was updated successfully, but these errors were encountered:
Currently, we know that the DLRM cuda libraries written by NVIDIA can run on V100 (compute capability 7.0) and A100 GPUs (compute capability 8.0). If we move to a newer GPU, such as RTX 3090 (compute capability 8.6), would the existing code work there? We might also need to change the nvcc flags for compilation.
The text was updated successfully, but these errors were encountered: