Drop MLX: 5.13x/2.25x slower than Core ML fp16 with no size win

mlx/NOTES.md

@@ -0,0 +1,40 @@
+# 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, 3 warmup runs + 15 timed runs,
+median, identical input shapes allocated from each model's own schema):
+
+| model | Core ML fp16 | MLX fp16 | MLX penalty |
+|---|---|---|---|
+| rfdetr-nano detector | **8.4 ms** | 43.1 ms | **5.13x slower** |
+| rfdetr-nano segmentation | **31.4 ms** | 70.5 ms | **2.25x slower** |
+
+Both backends delegate fully and return finite output; MLX is simply slower.
+
+**There is no size argument either.** The MLX and Core ML builds are within 1 MB
+of each other (detector 55.7 vs 54.7 MB, segmentation 62.0 vs 62.4 MB), so
+unlike CLIP and distiluse, where MLX int8 ships precisely because it is half the
+size of the Core ML build, there is nothing here to trade the latency against.
+
+## Why RF-DETR is not a special case
+
+MLX runs on the Metal GPU; Core ML lowers to the ANE, which is purpose-built for
+this kind of network. MLX was measured behind Core ML on every convolutional
+vision model tested, by 2.25x to 19.6x. Its real advantage is transformer
+workloads, where weight-only quantization shrinks the artifact substantially:
+CLIP vision is 97% `nn.Linear` and goes 351 MB to 98 MB, and distiluse goes
+541 MB to 140 MB. RF-DETR is a hybrid conv/transformer model and lands on the
+wrong side of that line.
+
+## A correction to the earlier figure
+
+An earlier note put the RF-DETR MLX penalty at about 33% (Core ML 51 ms vs MLX
+68 ms). That comparison is superseded and should not be quoted: it was measured
+on an iPhone SE 3rd gen, and against the Core ML **int8** build that has since
+been replaced by fp16, which is roughly 4x faster. The table above remeasures
+both backends at the same precision on the same device.

mlx/config.json

@@ -1,62 +0,0 @@
-{
-  "$schema": "https://huggingface.co/software-mansion/react-native-executorch-spec/resolve/main/config.schema.json",
-  "model": "rfdetr_nano_segmentation",
-  "family": "rfdetr",
-  "capabilities": [
-    "instance-segmentation"
-  ],
-  "backend": "mlx",
-  "license": "apache-2.0",
-  "variants": [
-    {
-      "file": "rfdetr_nano_seg_mlx_fp16.pte",
-      "precision": "fp16",
-      "quantized": false,
-      "default": true,
-      "methods": {
-        "forward": {
-          "inputs": [
-            {
-              "shape": [
-                1,
-                3,
-                312,
-                312
-              ],
-              "dtype": "float32"
-            }
-          ],
-          "outputs": [
-            {
-              "shape": [
-                100,
-                4
-              ],
-              "dtype": "float32"
-            },
-            {
-              "shape": [
-                100
-              ],
-              "dtype": "float32"
-            },
-            {
-              "shape": [
-                100
-              ],
-              "dtype": "float32"
-            },
-            {
-              "shape": [
-                100,
-                78,
-                78
-              ],
-              "dtype": "float32"
-            }
-          ]
-        }
-      }
-    }
-  ]
-}

mlx/rfdetr_nano_seg_mlx_fp16.pte

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:fa12474eca909e436cc64799f7d5c67508138467d15481fd3499b89920e4b753
-size 62006528