I think this usage pattern will be greatly appreciated
BTW as an optimization I would use Task scalars (they are send in the background, you can relativity easily get the latest value etc.) do you also need it to be atomic ?

Yes it's atomic, Okay I will research more about the Task scalars

So in theory 1,60,-1 should work as a size for Triton, Are you getting an error?
(BTW: if you were to manually run the model inference I'm assuming you would have created a 3d matrix where the dims are 1,60,<batch_size>, is that correct?

Unfortunately it's not working, in (1,60,1) dimension I think there is no batch size information there. The input data shape is like this:
[[1,...,60]]. I need the first dimension to be dynamic so I can send more time series data at once: [[1,..,60], [61,..,120], [121,..,180]] <- like that