Skip to main content
Ctrl+K
onnx-light-cpu  documentation - Home onnx-light-cpu  documentation - Home
  • Getting Started
  • Design
  • ByOp
  • API reference
  • Examples
  • Next Steps
  • GitHub
  • Getting Started
  • Design
  • ByOp
  • API reference
  • Examples
  • Next Steps
  • GitHub

Section Navigation

  • Abs (CPU)
  • Add (CPU)
  • And (CPU)
  • Attention (CPU)
  • BatchNormalization (CPU)
  • BitShift (CPU)
  • BitwiseAnd (CPU)
  • BitwiseOr (CPU)
  • BitwiseXor (CPU)
  • Cast (CPU)
  • Concat (CPU)
  • Div (CPU)
  • Equal (CPU)
  • Equal (CPU)
  • Equal (CPU)
  • Equal (CPU)
  • Exp (CPU)
  • Gather (CPU)
  • Gemm (CPU)
  • Greater (CPU)
  • GreaterOrEqual (CPU)
  • GroupNormalization (CPU)
  • InstanceNormalization (CPU)
  • LayerNormalization (CPU)
  • Less (CPU)
  • LessOrEqual (CPU)
  • LinearAttention (CPU)
  • Log (CPU)
  • LpNormalization (CPU)
  • MatMul (CPU)
  • MatMulInteger (CPU)
  • Max (CPU)
  • Mean (CPU)
  • MeanVarianceNormalization (CPU)
  • Min (CPU)
  • Mod (CPU)
  • Mul (CPU)
  • NonZero (CPU)
  • Not (CPU)
  • Or (CPU)
  • PRelu (CPU)
  • Pow (CPU)
  • QLinearMatMul (CPU)
  • RMSNormalization (CPU)
  • ReduceSum (CPU)
  • RotaryEmbedding (CPU)
  • ScatterND (CPU)
  • Sigmoid (CPU)
  • SimplifiedLayerNormalization (CPU)
  • Slice (CPU)
  • Softmax (CPU)
  • Split (CPU)
  • Sub (CPU)
  • Sum (CPU)
  • SwiGLU (CPU)
  • Tanh (CPU)
  • Xor (CPU)
  • TreeEnsemble (CPU)
  • BiasGelu (CPU)
  • CDist (CPU)
  • CausalConvWithState (CPU)
  • GatherBlockQuantized (CPU)
  • GroupQueryAttention (CPU)
  • LinearAttention (CPU)
  • MatMulNBits (CPU)
  • MoE (CPU)
  • MultiHeadAttention (CPU)
  • QMoE (CPU)
  • RotaryEmbedding (CPU)
  • SkipLayerNormalization (CPU)
  • SkipSimplifiedLayerNormalization (CPU)
  • ByOp
  • RotaryEmbedding (CPU)

RotaryEmbedding (CPU)#

  • Domain: ai.onnx

  • Device: CPU

  • Kernel: onnx_light_cpu::RotaryEmbedding

  • Supported types: FLOAT, FLOAT16, BFLOAT16

  • Opset bounds: since opset 23

Operator support#

Capability

Implementation

Shape inference

onnx_light_cpu::ComputeShapeRotaryEmbedding()

Peak memory

onnx_light_cpu::ComputePeakMemoryRotaryEmbedding()

Fusion patterns

None

Gradient

Not provided

previous

ReduceSum (CPU)

next

ScatterND (CPU)

On this page
  • Operator support
Show Source

Last updated on Oct 26, 2026.