File size: 2,097 Bytes
c70608c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 | # 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 deeplab-v3 (mobilenet_v3_large):
| backend | precision | median latency |
|---|---|---|
| Core ML | fp32 (precision matched) | 32.7 ms |
| MLX | fp32 (only precision offered) | 95.1 ms |
| Core ML | fp16 (shipped default) | **24.9 ms** |
MLX is **2.9x slower than Core ML at matched precision**, and **3.8x 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.
## 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.
|