pub fn w4a16_gemv_sw(
gpu: &dyn GpuBackend,
kernel: KernelHandle,
input: DevicePtr,
weight: &QuantizedWeight,
output: DevicePtr,
n: u32,
k: u32,
stream: u64,
) -> Result<()>Expand description
2026-09-25: Single-warp-per-output W4A16 GEMV for M=1, grid ceil(N/8).