Dynamic Shape Inference And Handling For Variable Length Inputs

via “multi-input/output model inference with dynamic shapes”

Cross-platform ONNX inference for mobile devices.

Unique: Implements shape inference at runtime by traversing the computation graph and applying shape propagation rules for each operator, enabling flexible input shapes without model recompilation. This is more flexible than TensorFlow Lite's approach, which requires fixed shapes or explicit shape specification.

vs others: More flexible than TensorFlow Lite because it supports arbitrary dynamic shapes without requiring model rebuilding; more efficient than PyTorch Mobile because shape inference is optimized for mobile devices with limited memory.

via “dynamic shape handling and symbolic dimension inference”

Cross-platform ML inference accelerator — runs ONNX models on any hardware with optimizations.

Unique: Implements symbolic dimension tracking (onnxruntime/core/graph/graph_utils.h) where tensor dimensions are represented as symbolic expressions (e.g., batch_size * seq_len) rather than fixed integers. Shape inference propagates these expressions through the graph, computing output shapes as functions of input dimensions. At runtime, symbolic variables are bound to actual values, enabling dynamic shape handling.

vs others: More flexible than TensorFlow's static shape model (which requires fixed shapes or explicit dynamic shape handling) and more efficient than PyTorch's dynamic shape handling (which recompiles the graph for each shape) because ORT infers shapes statically and binds them at runtime.

via “shape-polymorphic-tracing-and-compilation”

Google's numerical computing library — autodiff, JIT, vectorization, NumPy API for ML research.

Unique: JAX's shape polymorphism is integrated into jit — users can specify abstract shapes and jit automatically generates code that works for multiple concrete shapes. This is achieved through a tracer system that uses symbolic shapes during compilation and generates XLA code with runtime shape checks.

vs others: More efficient than recompiling for each shape because code is generated once; more flexible than static shape systems because shapes can vary at runtime

via “dynamic shape inference and handling for variable-length inputs”

OpenVINO™ is an open source toolkit for optimizing and deploying AI inference

Unique: Implements shape inference logic that propagates dynamic shapes through the graph, enabling inference with variable-length inputs without recompilation. The shape inference engine handles both static and dynamic dimensions, adapting to input variations at runtime.

vs others: Provides more flexible dynamic shape support than TensorFlow's static graph model and better shape inference than ONNX Runtime's limited dynamic shape support.