library_name: kernels
license: apache-2.0
tags:
- kernel
- webgpu
- wgsl
com.microsoft.Gelu
com.microsoft · ONNX Runtime contrib operator · contrib since_version 1
Description
Applies the Gaussian Error Linear Unit (GELU) activation elementwise: Y = 0.5 * X * (1 + erf(X / sqrt(2))). The output has the same shape as the input. Float16 and float32 are supported; the schema's double and bfloat16 types are not.
See the ONNX Runtime Gelu contrib-operator spec for the reference semantics.
Inputs
| Name | Bind key | Logical dtype | Rank | Shape | Description | Presence |
|---|---|---|---|---|---|---|
X |
X |
T |
— | — | Values transformed elementwise by the exact GELU activation. | required |
Outputs
| Name | Bind key | Logical dtype | Rank | Shape | Description | Presence |
|---|---|---|---|---|---|---|
Y |
Y |
T |
same as X |
same as X |
Output tensor after applying GELU; same shape as the input. | required |
Type constraints
| Variable | Allowed dtypes |
|---|---|
T |
float32, float16 |
Files
metadata.json— kernel metadata (id, digests, provenance)manifest.json— the op contract (source of truth)test.json— correctness casesbench.json— benchmark + tuning caseselementwise-bias-gelu.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/com.microsoft.Gelu", { version: 1 });
const { Y } = await kernel({ X: { data: XData, shape: [] } });