Category B: Binary Bundling
Embed ML models directly into Rust binaries for zero-dependency deployment.
Recipes
| Recipe | Description | Status |
|---|---|---|
| Bundle Static Model | Embed model with include_bytes!() | Verified |
| Bundle Quantized Model | Reduce model size with quantization | Verified |
| Bundle Encrypted Model | Protect model weights | Verified |
| Static Binary Embedding | Full static linking | Verified |
| Q4 Quantization | 4-bit quantization | Verified |
| Signed Models | Cryptographic signing | Verified |
| Lambda Package | AWS Lambda deployment | Verified |
Learning Objectives
- Embed models using
include_bytes!()macro - Reduce binary size with quantization
- Protect intellectual property with encryption
- Create single-binary deployments
Toyota Way: Muda (Waste Elimination)
Bundling eliminates external dependencies, reducing deployment complexity and potential failure points.