# FastVLM on iPhone and iOS

Use Apple’s official iOS/macOS demo project as the starting point for an on-device FastVLM application.

Canonical URL: https://aiidelist.com/blog/fastvlm-ios-coreml

Language: en

Published: 2026-09-24

Updated: 2026-09-24

FASTVLM GUIDE · iPhone

Use Apple’s official iOS/macOS demo project as the starting point for an on-device FastVLM application.

[All FastVLM guides](/fastvlm#guides)

## Before you run

The official app requires macOS 15.2+ or iOS 18.2+ and a compatible Xcode toolchain. Download the model before building. Use an empty model folder for the initial download; keep any existing custom model elsewhere before changing variants.

### Check your first result

Build and run in Xcode. Select the target device, configure signing for iPhone, then ask a question about an image or camera frame. Check the generated answer and the app’s TTFT display.

[Stage2 / Stage3 and Apple Silicon files →](/blog/fastvlm-1-5b-stage3#stage3)

Source checked September 12, 2026. Runtime performance depends on your device.

## Step by step

1. Use Apple’s app project rather than assuming a system-level FastVLM API.
2. Follow the repository’s model-download instructions.
3. Open and build the project with a compatible Xcode toolchain.
4. Measure memory and latency on every target device.

## Example

iPhone

```text
git clone https://github.com/apple-aiml-research/ml-fastvlm.git
cd ml-fastvlm
chmod +x app/get_pretrained_mlx_model.sh
app/get_pretrained_mlx_model.sh --model 0.5b --dest app/FastVLM/model
open app/FastVLM/FastVLM.xcodeproj
# Select an iOS 18.2+ device, configure signing, then Build and Run.
```

### Verify against the primary source

APIs, model files and dependency versions can change. Treat this page as an implementation map, then confirm the current command and license in the linked official documentation.

[Open official source](https://github.com/apple/ml-fastvlm/tree/main/app/FastVLM)

## FastVLM with Core ML

FASTVLM GUIDE · Core ML

Review Apple’s released Core ML resources and choose quantization deliberately for your target hardware.

[All FastVLM guides](/fastvlm#guides)

## Step by step

1. Open the official FastVLM collection and identify Core ML assets.
2. Match the model size and quantization to the target device.
3. Integrate the model package in a small test app first.
4. Benchmark warm and cold start separately.

## Example

Core ML

```text
// Add the downloaded .mlpackage to your Xcode target
// and let Xcode generate the typed model interface.
```

### Verify against the primary source

APIs, model files and dependency versions can change. Treat this page as an implementation map, then confirm the current command and license in the linked official documentation.

[Open official source](https://huggingface.co/collections/apple/fastvlm-68ca64a85e3c2a03f32a7098)