Function silu_mul_quant_fp8 

pub fn silu_mul_quant_fp8(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    gate: DevicePtr,
    up: DevicePtr,
    out_fp8: DevicePtr,
    a_scale: DevicePtr,
    out_bf16: DevicePtr,
    m: u32,
    k: u32,
    stream: u64,
) -> Result<()>

2026-09-25: SiLU(gate) * up quantized to FP8 E4M3 with one scale per 128-element group of each row, in one kernel (silu_mul_quant_fp8); the FP8 MoE prefill uses it before the down projection (moe/forward_prefill_fp8.rs).

out_bf16 may be DevicePtr::NULL; when it is not, the kernel also writes the post-SiLU BF16 values there. The caller must ensure k % 128 == 0 and k / 128 <= 16 (the kernel’s SILU_QUANT_MAX_GROUPS).