Face Swap SDK for Live Streaming Apps

Amigo AI SDK lets you add real-time face swap to any iOS live streaming app. It runs entirely on-device using CoreML at 30fps with 512px output — no server round-trips, no cloud costs, no added latency. The per-frame API accepts a CVPixelBuffer and returns a face-swapped CIImage, so it drops into any RTMP, HLS, or custom streaming pipeline. Pricing is $0.01 per session with 1,000 free every month.

How do I add face swap to my streaming app?

The SDK's processFrame() API is designed for real-time pipelines. You capture a camera frame, pass it through the SDK, and send the result to your streaming encoder. The entire pipeline runs on-device.

import AmigoFaceSwapSDK

// Initialize once at stream start
try await AmigoFaceSwap.initialize(apiKey: "your-key")
let latent = try await AmigoFaceSwap.enrollFace(from: facePhoto)

// In your camera frame callback:
func captureOutput(_ output: AVCaptureOutput,
    didOutput sampleBuffer: CMSampleBuffer, ...) {
  guard let pixelBuffer = sampleBuffer.imageBuffer else { return }
  if let swapped = try? AmigoFaceSwap.processFrame(
    pixelBuffer, using: latent
  ) {
    rtmpEncoder.encode(swapped) // Send to RTMP/HLS
  }
}

Why on-device face swap for streaming?

What performance can I expect?

Frequently asked questions