Very long compile times for moderately sized matmuls

TL;DR

When I construct an OpenVINO model consisting of just a single, moderately sized square matmul operation (in addition to the parameter and result ops), in order to launch on the NPU, the compile step takes really long.

The problem

I construct the model as follows:

def make_model (n):
    dtype = ov.Type ('float16')
    size = (n, n)

    A = ops.parameter (size, dtype, name = "A")
    B = ops.parameter (size, dtype, name = "B")
    C = ops.matmul (A, B, False, False)
    res = ops.result (C)

    return ov.Model ([res], [A, B], "matmul")

And I compile it as follows:

compiled_model = core.compile_model (make_model (11264), "NPU")

Some numbers:

• n=11264 (i.e. matmul with 11264x11264 matrices) takes 10 minutes to compile
• n=12288 takes 23 minutes
• n=13312 takes 114 minutes

After compilation, inference (i.e. running the matmul on random inputs) is as quick as I'd expect it to be. Subsequent runs for a single matrix size spend no time on compilation, probably since they are satisfied by the NPU model cache.

I've tried raising the thread limit for the compiler, but it still runs on a single thread.

System info

• OpenVINO 2025.2.0 and OpenVINO 2025.3.0.dev20250729 (nightly)
• Linux NPU Driver v1.19.0, which I understand includes the NPU Compiler 2025.24
• Ubuntu 24.10 Oracular, with Linux 6.11.29
• NPU device architecture: 4000

Reproduce the Python model from the issue with 11264, 12288, and 13312 matrix sizes, compiling each with core.compile_model(..., "NPU"). Measure compile time and thread usage, then trace the NPU compiler path responsible for matmul compilation. Done means moderately sized matmuls compile substantially faster without regressing inference or model-cache behavior.