ai.onnx.GatherElements

ai.onnx · standard ONNX operator · ONNX opset ≥ 13

Description

Gathers values from the data tensor using index positions specified by indices, both of the same rank. For each element in indices, the corresponding output value is read from data by substituting that index along the chosen axis while keeping all other coordinates the same. The output shape equals the shape of indices.

See the ONNX GatherElements spec for the reference semantics.

Inputs

Outputs

Attributes

Default values (overridable per request):

Type constraints

Files

• metadata.json — kernel metadata (id, digests, provenance)
• manifest.json — the op contract (source of truth)
• test.json — correctness cases
• bench.json — benchmark + tuning cases
• gather-elements-rank2-vec4.wgsl.jinja
• gather-elements.wgsl.jinja

Use with @huggingface/kernels

The loader derives every required output's shape and logical dtype from the manifest contract and this call. It then allocates the result tensors automatically.

The version: 1 option selects the published kernel contract; it is independent of any operator opset, contrib since_version, or model version.

Replace each *Data placeholder with a typed array containing the corresponding input data.

import { getKernel } from "@huggingface/kernels";

const kernel = await getKernel("webgpu-kernels/ai.onnx.GatherElements", { version: 1 });
const { output } = await kernel({
  data: { data: dataData, shape: [2, 3] },
  indices: { data: indicesData, shape: [1, 2] },
});