webnn-graph v0.3.0

[0.3.0] - 2026-05-04

Added

• Dynamic dimension representation in AST and WG grammar via dyn("name", maxSize) input dims.
• ONNX conversion support for preserving unresolved dynamic input dimensions in graph metadata.
• New convert-onnx CLI flag: --experimental-dynamic-inputs (opt-in dynamic input preservation).
• ONNX converter/operator support improvements:
  • ScatterND
  • Where, Equal, comparison operators, Cos, Sin, TriLu, Tile
  • ConstantOfShape, Range, and additional constant-folding evaluators
• Built-in constant folding in webnn-graph (--optimize) to reduce dynamic-shape plumbing.
• Global debug switch for converter diagnostics (--debug).
• Pre-commit hook setup script and make-based local checks.

Changed

• ONNX conversion now supports static-lowering + dynamic metadata workflows in one pipeline.
• Graph parser/serializer now support richer values (including object literals in options).
• JS/HTML emitters and visualizer now render mixed static/dynamic shapes.
• Docs expanded and corrected:
  • ONNX lowering behavior
  • Dynamic dimension guidance
  • SmolLM-135M conversion example from Hugging Face

Fixed

• Multiple ONNX conversion correctness fixes, including:
  • dynamic reshape/expand conversion edge cases
  • shape inflation prevention and post-conversion shape tracking
  • Unsqueeze v14 handling
  • identifier sanitization robustness (including $ prefixes)
  • clippy/robustness cleanup across converter and shape inference

Compatibility

• Existing static graphs remain supported.
• Validator/serializer support both graph versions v1 and v2.
• Dynamic input metadata is experimental and must be enabled with
  --experimental-dynamic-inputs.

0.2.1

Full Changelog: v0.2.0...0.2.1

v0.2.0: WebNN Text Format and Binary Weights

What's New

Features

• WebNN text format (.webnn) is now the primary format - 10x smaller than JSON, human-readable
• Binary weights support - Separate weight data from graph structure
  • extract-weights: Extract weights from JSON to binary format with manifest
  • inline-weights: Inline weights back into JSON for round-trip conversion
  • pack-weights, unpack-weights, create-manifest: Low-level weight management
• Improved serialization - JSON to .webnn conversion no longer includes inline bytes

Improvements

• Enhanced documentation with complete workflow examples
• Better separation of concerns: graph structure, weights, and input data
• Full round-trip support: .webnn ↔ JSON ↔ .webnn
• Clippy warnings fixed and code formatted
• Version bumped to 0.2.0 in Cargo.toml

Technical Details

• All commands now accept both .webnn and .json formats (auto-detected)
• Weight extraction achieves 93% size reduction for large models
• Complete test coverage across all modules (50 tests)
• All tests passing, clippy clean
• Ready for crates.io publication

Full Changelog: v0.1.0...v0.2.0

v0.1.0 - Initial Release

Initial release of webnn-graph, a DSL for webnn.

• Basic parsing of .webnn text files or JSON files.
• Javascript code generator