Function nvfp4_proj_mx 

pub fn nvfp4_proj_mx(
    gpu: &dyn GpuBackend,
    input: DevicePtr,
    weight: &QuantizedWeight,
    output: DevicePtr,
    m: u32,
    n: u32,
    k: u32,
    stream: u64,
) -> Result<()>

2026-09-30: [m, k] x [n, k]^T with NVFP4 activations for any m: per chunk of the widest prepared entry, the row quantizer and then the mx GEMV. Every mx entry sums a row the same way and the quantizer is per row, so a row’s output bits do not depend on m or on the other rows. Errors when the kernels were not prepared or the shape does not fit; it never falls back to another format. For a fixed --activation-quantization only.