The following table lists the ai.onnx operators supported by each of the available backends.
See Compatibility for a list of the supported platforms.
Operator | Cpu Backend | Wasm Backend | WebGl Backend |
---|---|---|---|
Abs | ✔️ | ✔️ | |
Acos | ✔️ | ✔️ | |
Add | ✔️ | ✔️ | ✔️ |
And | ✔️ | ✔️ | ✔️ |
ArgMax | ✔️ | ||
Asin | ✔️ | ✔️ | |
Atan | ✔️ | ✔️ | |
AveragePool | ✔️ | ✔️ | ✔️ |
BatchNormalization | ✔️ | ✔️ | ✔️ |
Ceil | ✔️ | ✔️ | |
Concat | ✔️ | ✔️ | |
Constant | ✔️ | ✔️ | ✔️ |
Conv | ✔️ | ✔️ | ✔️ |
Cos | ✔️ | ✔️ | |
Div | ✔️ | ✔️ | ✔️ |
Dropout | ✔️ | ✔️ | ✔️ |
Equal | ✔️ | ||
Exp | ✔️ | ✔️ | |
Floor | ✔️ | ✔️ | |
Flatten | ✔️ | ✔️ | |
Gather | ✔️ | ✔️ | |
Gemm | ✔️ | ✔️ | |
GlobalAveragePool | ✔️ | ✔️ | ✔️ |
GlobalMaxPool | ✔️ | ✔️ | ✔️ |
Greater | ✔️ | ||
Identity | ✔️ | ✔️ | ✔️ |
ImageScaler | ✔️ | ✔️ | |
InstanceNormalization | ✔️ | ✔️ | |
LeakyRelu | ✔️ | ✔️ | |
Less | ✔️ | ||
Log | ✔️ | ✔️ | |
LRN | ✔️ | ||
MatMul | ✔️ | ✔️ | ✔️ |
MaxPool | ✔️ | ✔️ | ✔️ |
Mul | ✔️ | ✔️ | ✔️ |
Neg | ✔️ | ✔️ | |
Not | ✔️ | ||
Or | ✔️ | ✔️ | ✔️ |
Pad | ✔️ | ||
Pow | ✔️ | ||
PRelu | ✔️ | ✔️ | |
ReduceLogSum | ✔️ | ✔️ | |
ReduceMax | ✔️ | ✔️ | |
ReduceMean | ✔️ | ✔️ | |
ReduceMin | ✔️ | ||
ReduceProd | ✔️ | ||
ReduceSum | ✔️ | ||
ReduceSumSquare | ✔️ | ||
Relu | ✔️ | ✔️ | |
Reshape | ✔️ | ✔️ | |
Sigmoid | ✔️ | ✔️ | |
Sin | ✔️ | ✔️ | |
Slice | ✔️ | ✔️ | |
Softmax | ✔️ | ✔️ | ✔️ |
Split | ✔️ | ||
Sqrt | ✔️ | ✔️ | |
Sub | ✔️ | ✔️ | ✔️ |
Sum | ✔️ | ✔️ | ✔️ |
Tan | ✔️ | ✔️ | |
Tanh | ✔️ | ✔️ | |
Tile | ✔️ | ✔️ | |
Transpose | ✔️ | ✔️ | |
Unsqueeze | ✔️ | ✔️ | |
Xor | ✔️ | ✔️ | ✔️ |