XPU Lens · Free
Look inside any AI model.
Open a model and explore it block by block, down to every layer: what goes in and out, its weights, and what it costs in multiply-accumulates.
- ONNX
- TFLite
- Keras
- PyTorch
- TensorFlow
Or try a sample
Inputs
Outputs
MACs are estimated from layer shapes, for convolutions, dense layers and matrix products; other layers count as none.