Function glm_hc_pre 

pub fn glm_hc_pre(
    gpu: &dyn GpuBackend,
    kernels: &Glm5NextMhcKernels,
    streams: DevicePtr,
    w: &Glm5NextMhcSiteWeights,
    y_out: DevicePtr,
    post_out: DevicePtr,
    comb_out: DevicePtr,
    num_tokens: u32,
    hidden_size: u32,
    hc_mult: u32,
    sinkhorn_iters: u32,
    norm_eps: f32,
    hc_eps: f32,
    stream: u64,
) -> Result<()>

2026-09-25: Collapse each token’s hc_mult FP32 streams to one BF16 row in y_out, and write this site’s post and comb, by launching hc_mix (or hc_mix_bf16) then hc_finish. streams is read, not written, so glm_hc_post can take it as its residual. Errors when num_tokens exceeds mhc_mix_max_tokens().