Native Broadcast op-node — tile a vector across a larger output by index remap
(out[i] = input[i % in_len]). This is RMSNorm/bias scale-fanout: a per-feature
vector scale[d] expanded across tokens so out[t*d + j] = scale[j]. No arithmetic,
one invocation per output element.
Native GatherDequant op-node — on-the-fly ternary weight dequant producing an f32
tensor that a downstream MatMul consumes (the {GatherDequant → MatMul} split, plan
§2 / P4b). This is the LLM weight-decompression primitive: BitNet-style 2-bit ternary
codes unpacked to {0, +1, −1} and scaled per row.
Native Reduce op-node — the first compute-graph node the WGSL backend emits
from scratch (Phase 1’s gemm/gemv/fft delegate to legacy emitters; there is no
legacy standalone reduce, so this is a real graph-template lowering).
Native ScatterAccum op-node — scatter src[p] into output[idx[p]] with an associative
accumulate (plan §2 / P7): the SPH density/force deposit and N-body short-range gather.
Native Stencil op-node — local neighbourhood maps over a 1-D grid (plan §2 / P7). The
seed kind is the Laplacianout[i] = in[i-1] − 2·in[i] + in[i+1] with zero (Dirichlet)
boundaries — the discrete Poisson operator under fluid pressure-solve / diffusion, and the
Reduce-free half of a stencil sweep.