AOZMH

Thanks for the follow-up, I'll provide my code snippet and the execution logs (time costs) shortly after!

> Roughly, we can know that T(cross+bi) : T(bi) = 3:1, ignoring the other process * Thanks for the theoretical analysis! That's a bit different from the case I encountered,...

> Hi @AOZMH, > Thanks for reporting. It is known that the (fast=False) mode is slow, however, the faiss index should be faster. Could you provide a code snippet? Thanks....

> Hey, how did you manually change biencoder to gpu? Could you share the snippets? You can simply revert [this](https://github.com/facebookresearch/BLINK/blob/master/blink/main_dense.py#L251) commented code to revert the transition to GPU (**# .to(device)**...

> I am facing the same memory issue you did. Could you elaborate your third point - On how to lower memory usage? I tried two ways as follow: 1....

BTW, I really hope the developers of BLINK can look into this issue to solve the faiss index problem I mentioned before, thanks in advance!

> I wonder if this is also a feasible solution - Splitting the candidate_encoding when you pass it to gpu and then concatenating the splittled scores and continuing with the...