Does encoding multiple sentences using model.encode, have a limit to no. of sentences accepted?

[ankitas3]

The SentenceTransformers Documentation says that multiple sentences are also accepted while encoding. Quoting: Even though we talk about sentence embeddings, you can use it also for shorter phrases as well as for longer texts with multiple sentences Is there any limit to length of array of sentences that is acceptable here?

[nreimers]

Depending on the machine, at some point you will run out of memory, as each vector needs ~768*4 bytes of memory

[ankitas3]

So there's no limit as such but has to be handled depending on the memory. Thanks @nreimers

[ankitas3]

@nreimers Is there any optimal number for array length maybe based on the system core CPU/GPU to have an idea around this.