geoffroyleconte

Fixing `obj` would require writing a function that computes `x' * H * x` without allocating (for all `H` types).

Oh I didn't know about this function. It would have to be added to SparseMatricesCOO. Also I would keep the existing behaviour in case H is a LinearOperator.

Yes for SparseMatricesCOO. For LinearOperators this is complicated, I would keep the existing code even though it allocates.

Not everything is compatible. I managed to create some QuadraticModels and use them with CUDA using for example https://github.com/JuliaSmoothOptimizers/RipQP.jl/issues/289#issuecomment-2081660874 . However, `SlackModel` in NLPModelsModifiers does not work. I had to...

It looks good, even though i'm not sure to understand what this does: https://github.com/JuliaSmoothOptimizers/NLPModelsModifiers.jl/blob/d90bd8e564542d2654ac83596f9a3c1b513cc197/src/slack-model.jl#L162 But what is important is for this function to be implemented on the GPU: https://github.com/JuliaSmoothOptimizers/QuadraticModels.jl/blob/0c05ef84f76410ed818a469371f2f8aff6336e48/src/qpmodel.jl#L548-L572 because...

Is this issue fixed, or did you mean something more elaborated?

OK, I can add a tutorial

https://github.com/JuliaSmoothOptimizers/QuadraticModels.jl/pull/75