I noticed that when I use API DeepSeek model is much faster than in UI. It would be good if you would have switch or something to increase speed by using API.
The infrastructure that powers the API and the App are the same, so tokens per second of generation should be identical. Are you testing these at the same time of day? If you continue to have issues, please email [email protected].
Log in to comment and vote
Comments1
JP
Apr 7, 2025
The infrastructure that powers the API and the App are the same, so tokens per second of generation should be identical. Are you testing these at the same time of day? If you continue to have issues, please email [email protected].