Files
carmackandHanzo Dev 44bdc61e5d hanzo-ml 0.11.91: metal(qmatmul) batched small-m matvec for quantized weights
Route m<=metal_matvec_max() (default 8) quantized matmuls through a single
batched ne11=m matvec dispatch instead of a per-column m-loop, cutting
verify-round latency in speculative decode. Dense F16/BF16/F32 weights stay on
the GEMM path: the batched matvec family indexes src0/src1 by block/element
counts and would mis-stride and mis-grid dense weights, so fwd() dtype-gates the
batched route to the quantized families only.

Adds a metal-kernels F16-GEMM-vs-CPU test and an fwd()-level F16 m=4 routing test.
2026-07-22 02:29:35 -07:00
..

hanzo-metal-kernels

This crate contains Metal kernels used from hanzo-ml.