Function w4a16_gemm 

pub fn w4a16_gemm(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    input: DevicePtr,
    weight: &QuantizedWeight,
    output: DevicePtr,
    m: u32,
    n: u32,
    k: u32,
    stream: u64,
) -> Result<()>

2026-09-29: W4A16 GEMM, C = A @ dequant(B): A [M, K] BF16, B NVFP4 (packed E2M1, FP8 block scales, FP32 per-tensor scale), C [M, N] BF16. Every launcher here sizes grid.x from the kernel’s published N tile (n_tile_blocks).

Also launches w4a16_gemm_t_k64_n64_p3, which takes the same arguments.