I used a fast-start demo that took 30 seconds to complete on a 4090Ti.
But I use the web service interface processing time only needs 200ms, read the code, the core reasoning method and parameters are basically the same, could you tell me why, thank you
I used a fast-start demo that took 30 seconds to complete on a 4090Ti. But I use the web service interface processing time only needs 200ms, read the code, the core reasoning method and parameters are basically the same, could you tell me why, thank you