Sure thing, happy to share. That potential root cause makes sense. I bet magnitude will close the prefill gap then. llama.cpp had that same matmul gap (pre-fill operations) for the M5/A19 and newer silicon until that…
To OP: great work on the release! I am generally interested in this kind of optimization work. Related to the post above: Similar results here M5 Max running Qwen3.8 UD-Q6-K-XL with zlab’s Dflash2 as the drafter. Both…
Thanks for all of your exploration in public Simon. Commenting because the fix I proposed was merged in roughly 49 commits after the PrismML Fork. The “tensor API is not supported” warning occurred because llama.cpp’s…
Sure thing, happy to share. That potential root cause makes sense. I bet magnitude will close the prefill gap then. llama.cpp had that same matmul gap (pre-fill operations) for the M5/A19 and newer silicon until that…
To OP: great work on the release! I am generally interested in this kind of optimization work. Related to the post above: Similar results here M5 Max running Qwen3.8 UD-Q6-K-XL with zlab’s Dflash2 as the drafter. Both…
Thanks for all of your exploration in public Simon. Commenting because the fix I proposed was merged in roughly 49 commits after the PrismML Fork. The “tensor API is not supported” warning occurred because llama.cpp’s…