pub fn launch_k3_moe_e8m0_ptrtable(
gpu: &dyn GpuBackend,
kernel: KernelHandle,
a: DevicePtr,
packed: &[QuantizedWeight],
c: DevicePtr,
expert_offsets: DevicePtr,
sorted_token_ids: DevicePtr,
num_experts: u32,
n_out: u32,
k: u32,
stream: u64,
) -> Result<()>Expand description
2026-09-25: One grouped E8M0 GEMM, C[M, N] = A[M, K] @ B_packed[N, K/2],
with one pointer-table entry per expert in packed. The launch covers one
M tile per expert; gemm_rows gives every expert exactly one row.