| # MLX is not shipped for this model |
|
|
| The MLX artifacts for this model were removed deliberately. The exporter in |
| `export-scripts` is kept so the decision can be re-tested against a future |
| ExecuTorch/MLX release, but no `.pte` or `config.json` is published here. |
|
|
| ## Why |
|
|
| Measured on device (iPhone 16, ExecuTorch 1.4.1, Release build, 3 warmup runs + |
| 15 timed runs, median), on selfie-segmentation: |
|
|
| | backend | precision | median latency | |
| |---|---|---| |
| | Core ML | fp32 (precision matched) | 4.5 ms | |
| | MLX | fp32 (only precision offered) | 5.1 ms | |
| | Core ML | fp16 (shipped default) | **0.3 ms** | |
|
|
| MLX is **1.1x slower than Core ML at matched precision**, and **17.0x slower than |
| the Core ML build that actually ships**. |
|
|
| These exporters only ever declared an MLX fp32 variant, so fp32-vs-fp32 is the |
| fair comparison and it is deliberately the conservative one: MLX still loses it. |
|
|
| **Caveat specific to this model.** At matched precision MLX is only 1.1x behind |
| Core ML, which is near parity and the closest result in this study. MLX is not |
| performing badly here. It is removed because Core ML fp16 is exceptional on this |
| model (0.3 ms, a 15x gain over its own fp32), and fp16 is what ships. If Core ML |
| fp16 ever regressed for this model, MLX would be a reasonable fallback. |
|
|
| ## Context |
|
|
| MLX was measured behind Core ML on every convolutional vision model tested. The |
| margin varies widely with architecture, so each model carries its own number |
| rather than a blanket figure. |
|
|
| Two things compound the gap. MLX weight quantization cannot shrink a conv model: |
| `quantize_model_` only rewrites `nn.Linear` and embedding modules, and |
| EfficientNet-V2-S holds just 6.0% of its parameters in its single `nn.Linear`, |
| so 4-bit quantization buys about 5% file size. Core ML by contrast reaches the |
| convolutions. |
|
|
| ## Scope |
|
|
| This concerns convolution-dominated vision models. It does **not** generalize: |
|
|
| - `rfdetr-nano` (conv + transformer) measured only **1.24x** behind Core ML. |
| - The `lfm2.5` text encoders measured **2.7-3.1x faster on MLX** than XNNPACK. |
|
|
| MLX suits matmul-heavy graphs. The right backend depends on model class, so this |
| file is not a statement about MLX in general. |
|
|
| ## Availability history |
|
|
| MLX for this model was only ever published on `main` / `v0.10.0`. No earlier tag |
| (`v0.9.0` and below) carried an MLX artifact for it, so there is no pinned |
| revision from which these files can still be fetched. They remain recoverable |
| from this repository's git history. |
|
|
