cutlass_nvfp4_proj_from_fp8

Function cutlass_nvfp4_proj_from_fp8 

Source
pub fn cutlass_nvfp4_proj_from_fp8(
    ctx: &ForwardContext<'_>,
    act: DevicePtr,
    fp8w: &Fp8Weight,
    out: DevicePtr,
    m: u32,
    n: u32,
    k: u32,
    stream: u64,
) -> Result<()>
Expand description

Native CUTLASS NVFP4 projection for FP8 checkpoint weights. The FP8 weight is dequantized to BF16 using the existing cache, then packed once into Atlas-transposed NVFP4 data/scales and reused for future calls.