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
Hi, when generating embeddings for chunks, CPU takes longer than GPU:
On CPU:
CPU times: user 6min 12s, sys: 3min 52s, total: 10min 4s
Wall time: 1min 43s
On GPU:
CPU times: user 3min 10s, sys: 9.06 s, total: 3min 19s
Wall time: 3min 18s
Note: I'm running it on a Mac with M1 Pro using "mps" device for GPU instead of "cuda" device. Also, I changed the embedding model to multilingual version of the same model: sentence-transformers/paraphrase-multilingual-mpnet-base-v2
In the example, with a RTX4090 GPU times are 1/10 of CPU times... why is my GPU taking almost double time than CPU?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hi, when generating embeddings for chunks, CPU takes longer than GPU:
On CPU:
CPU times: user 6min 12s, sys: 3min 52s, total: 10min 4s
Wall time: 1min 43s
On GPU:
CPU times: user 3min 10s, sys: 9.06 s, total: 3min 19s
Wall time: 3min 18s
Note: I'm running it on a Mac with M1 Pro using "mps" device for GPU instead of "cuda" device. Also, I changed the embedding model to multilingual version of the same model: sentence-transformers/paraphrase-multilingual-mpnet-base-v2
In the example, with a RTX4090 GPU times are 1/10 of CPU times... why is my GPU taking almost double time than CPU?
Beta Was this translation helpful? Give feedback.
All reactions