Category U: Format
Format recipes demonstrate the APR format ecosystem: importing models from external hubs, exporting to interoperable formats (SafeTensors, GGUF), cross-format conversion via the Rosetta engine, quantized conversion, publishing, and batch operations. These mirror the apr import, apr export, apr rosetta, apr convert, apr publish, and apr pull CLI subcommands.
Recipes
| # | Recipe | CLI Equivalent | Description |
|---|---|---|---|
| 1 | Import from HuggingFace | apr import hf://org/repo | Download and convert a HuggingFace model to .apr |
| 2 | Export to SafeTensors | apr export --format safetensors | Serialize an .apr model to SafeTensors format |
| 3 | Export to GGUF | apr export --format gguf | Serialize an .apr model to GGUF for llama.cpp |
| 4 | Rosetta Convert | apr rosetta convert | Cross-format conversion via the Rosetta engine |
| 5 | Rosetta Chain | apr rosetta chain | Multi-step conversion chain (e.g., ONNX -> APR -> GGUF) |
| 6 | Rosetta Verify | apr rosetta verify | Round-trip verification of format fidelity |
| 7 | Convert with Quantization | apr convert --quantize | Convert between formats with quantization applied |
| 8 | Publish to HuggingFace | apr publish | Push an .apr model to a HuggingFace repository |
| 9 | Pull and Cache | apr pull | Download models with local cache management |
| 10 | Batch Multi-Format Export | apr export --batch | Export a model to multiple formats in one pass |