Sub

Performs element-wise binary subtraction (with Numpy-style broadcasting support). This operator supports multidirectional (i.e., Numpy-style) broadcasting; for more details please check [the doc](Broadcasting.md). (Opset 14 change): Extend supported types to include uint8, int8, uint16, and int1…

Abstract Signature:

Sub(A: Tensor, B: Tensor)

PyTorch

API: torch.sub
Strategy: Direct Mapping
Official Docs ↗

JAX (Core)

API: jnp.subtract
Strategy: Direct Mapping
Official Docs ↗

NumPy

API: np.subtract
Strategy: Infix (-)
Official Docs ↗

Keras

API: keras.ops.subtract
Strategy: Direct Mapping
Official Docs ↗

TensorFlow

API: tf.math.subtract
Strategy: Direct Mapping
Official Docs ↗

Apple MLX

API: mx.subtract
Strategy: Direct Mapping
Official Docs ↗

NVIDIA SASS

API: FADD
Strategy: Direct Mapping

AMD RDNA

API: v_sub_f32
Strategy: Direct Mapping
Official Docs ↗

Flax NNX

API: optax.tree.sub
Strategy: Direct Mapping
Official Docs ↗