mlboydaisuke/YOLO-World-V2-S-CoreML

YOLO-World — Core ML

Open-Vocabulary Detection, 2024

Open-vocabulary detection. Type any text query. YOLO-World V2-S + CLIP ViT-B/32.

Core ML conversion of AILab-CVC/YOLO-World for on-device inference on iPhone, iPad and Mac. Converted with coremltools; the packages are stateless, so all sequencing and buffering lives in your Swift code.

Task	zero shot object detection
Upstream	AILab-CVC/YOLO-World
Packages	2
Download size	134 MB
Minimum iOS	17.0
Peak RAM	~600 MB

Files

File	Size	Compute units	SHA-256
yoloworld_detector.mlpackage.zip	23 MB	all	611d299ae74c83f9…
clip_text_encoder.mlpackage.zip	111 MB	cpuOnly	45770a743297e8c2…
Total	134 MB

compute_units is not a suggestion -- it is the configuration the conversion was verified against. Moving a package to a different compute unit can silently change the numerics (FP16 attention overflow) or crash on the GPU.

Download

hf download mlboydaisuke/coreml-zoo --include "yoloworld/*" --local-dir ./yoloworld
unzip './yoloworld/yoloworld/*.zip' -d ./yoloworld

Use in Swift

import CoreML

let config = MLModelConfiguration()
config.computeUnits = .all   // as converted — see the table above

// Unzip the .mlpackage, drop it into your Xcode target and Xcode compiles it
// at build time:
let model = try yoloworld_detector(configuration: config)

// ...or compile a downloaded .mlpackage at runtime:
let compiled = try await MLModel.compileModel(at: mlpackageURL)
let model = try MLModel(contentsOf: compiled, configuration: config)

This model is split into 2 Core ML packages that are driven in sequence from Swift. Load them one at a time, copy the outputs out of the MLMultiArray buffers and release each model before loading the next — two large Core ML models resident at once will OOM on an iPhone.

Demo

• Sample app — sample_apps/YOLOWorldDemo, a standalone SwiftUI project.
• Models Zoo — this model is downloadable and runnable inside the Models Zoo app on the App Store, no build required.

Conversion

• Pitfalls hit during conversion (FP16 overflow, ANE buffer limits, stride handling): docs/coreml_conversion_notes.md
• Model index: CoreML-Models

License

The conversion inherits the upstream license: GPL-3.0.

Credits

• Upstream authors: AILab-CVC/YOLO-World, 2024
• Core ML conversion: john-rocky (Daisuke Majima)