image generation model locals actually use

[sparrowshed]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at token cost.

[phasescope]

Thanks, that helps. The pattern around prefix caching has been pretty similar for me.

[mlxgarden]

Appreciate the detail. I ended up in about the same place using llama.cpp.

[kernelghost]

What worked for me was pretty simple. If you're on MLX, I'd confirm the version first and then reduce variables until the behavior is obvious. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.