pub fn gemv_gate_up(
gpu: &dyn GpuBackend,
gate: &MlxInt8Weight,
up: &MlxInt8Weight,
x: DevicePtr,
gate_y: DevicePtr,
up_y: DevicePtr,
stream: u64,
) -> Result<()>Expand description
Dual-output GEMV: gate_y = gate @ x and up_y = up @ x in one
kernel launch. Halves the x-side memory bandwidth and removes one
kernel-launch round-trip per FFN. Both projections must share
(out_features, in_features, group_size).