Upload folder using huggingface_hub

README.md

@@ -25,9 +25,10 @@ This is a native ML Program format CoreML model converted from RMBG-2.0 (BiRefNe
 ### Key Features
 
 - **Native CoreML ML Program format** - Full ANE compatibility
+- **INT8 weight quantization** - Reduced model size (233MB vs 461MB)
 - **No custom layers** - Uses only native CoreML operations
 - **1024x1024 input resolution** - High-quality segmentation
-- **Fast inference** - ~3 seconds on Apple Silicon with ANE
+- **Fast inference** - ~3-5 seconds on Apple Silicon
 
 ## Usage
 
@@ -78,6 +79,8 @@ let mask = output.featureValue(for: "output_3")?.multiArrayValue
 | Normalization | ImageNet (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) |
 | Output | 4 scales (output_0 to output_3), use output_3 for full resolution |
 | Format | ML Program (.mlpackage) |
+| Weight Quantization | INT8 (symmetric linear) |
+| Model Size | ~233 MB |
 | Minimum OS | macOS 13+ / iOS 16+ |
 | Compute Units | All (CPU, GPU, ANE) |
 
@@ -85,14 +88,23 @@ let mask = output.featureValue(for: "output_3")?.multiArrayValue
 
 Tested on Apple Silicon:
 
-| Device | Inference Time |
-|--------|---------------|
-| M1 Pro (ANE) | ~3s |
-| M1 Pro (GPU) | ~5s |
+| Device | Compute Units | Inference Time |
+|--------|--------------|---------------|
+| M1 Pro | .all (ANE) | ~5s |
+| M1 Pro | .cpuAndGPU | ~3s |
+
+> Note: ANE performance varies by device. The INT8 quantization improves memory efficiency and model size while maintaining quality.
 
 ## Files
 
-- `RMBG-2-native.mlpackage/` - CoreML ML Program model
+Two model versions are available:
+
+| File | Size | Description |
+|------|------|-------------|
+| `RMBG-2-native.mlpackage/` | 461 MB | Original FP32 model |
+| `RMBG-2-native-int8.mlpackage/` | 233 MB | INT8 quantized model (recommended) |
+
+**Recommended:** Use the INT8 version for better memory efficiency and smaller download size with equivalent quality.
 
 ## License
 

RMBG-2-native-int8.mlpackage/Data/com.apple.CoreML/model.mlmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef6031ea7ef7c8ae91217a74606208db7c82ec5adca112c5224db9ebe021947c
+size 3252446

RMBG-2-native-int8.mlpackage/Data/com.apple.CoreML/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d85cd30f732009782edd92614b1f0a740a82af5ca42191fde802dff466925467
+size 241333920

RMBG-2-native-int8.mlpackage/Manifest.json

@@ -0,0 +1,18 @@
+{
+    "fileFormatVersion": "1.0.0",
+    "itemInfoEntries": {
+        "1FA3F375-5BE6-43BE-BFEB-6B8490589024": {
+            "author": "com.apple.CoreML",
+            "description": "CoreML Model Specification",
+            "name": "model.mlmodel",
+            "path": "com.apple.CoreML/model.mlmodel"
+        },
+        "F71FD107-4CF6-4D1F-922B-CCA5072E55E6": {
+            "author": "com.apple.CoreML",
+            "description": "CoreML Model Weights",
+            "name": "weights",
+            "path": "com.apple.CoreML/weights"
+        }
+    },
+    "rootModelIdentifier": "1FA3F375-5BE6-43BE-BFEB-6B8490589024"
+}

config.json

@@ -11,6 +11,8 @@
   "outputs": ["output_0", "output_1", "output_2", "output_3"],
   "primary_output": "output_3",
   "format": "mlpackage",
+  "weight_quantization": "int8_symmetric_linear",
+  "model_size_mb": 233,
   "compute_units": "all",
   "minimum_deployment_target": {
     "macOS": "13.0",
@@ -18,5 +20,5 @@
   },
   "original_model": "briaai/RMBG-2.0",
   "license": "CC-BY-NC-4.0",
-  "version": "1.0.0"
+  "version": "1.1.0"
 }