optimum
optimum copied to clipboard
what the difference between decoder_model.onnx and decoder_with_past_model.onnx
Feature request
none
Motivation
none
Your contribution
none
Hi @akk-123,
The difference is that the decoder_with_past_model.onnx has the pre-computed key/values hidden-states as one of its inputs while the decoder_model.onnx has not. See here for more information.
@echarlaix The link is not available anymore, can you please update your last comment?
@ayansengupta17 It has been moved here.