Upload README.md with huggingface_hub

README.md

@@ -12,6 +12,52 @@ tags:
 library_name: coreml
 ---
 
+## Use it from Swift
+
+<!-- swift-usage-begin -->
+### Add the package
+
+`Package.swift`:
+
+```swift
+.package(url: "https://github.com/john-rocky/CoreML-LLM", branch: "main"),
+
+// In your target:
+.product(name: "CoreMLLLM", package: "CoreML-LLM"),
+```
+
+Platforms: iOS 18+ / macOS 15+.
+
+### Download + encode
+
+```swift
+import CoreMLLLM
+
+let modelsDir = try FileManager.default.url(
+    for: .applicationSupportDirectory, in: .userDomainMask,
+    appropriateFor: nil, create: true)
+
+let eg = try await EmbeddingGemma.downloadAndLoad(modelsDir: modelsDir)
+
+// 768-dim L2-normalised embedding
+let v = try eg.encode(text: "How do I list files in Swift?")
+// Matryoshka: cheap-to-truncate dims (768 / 512 / 256 / 128)
+let v256 = try eg.encode(text: "How do I list files in Swift?",
+                          dim: 256)
+
+// Task-prefixed (RAG document vs. query)
+let q = try eg.encode(text: "list files",
+                       task: .retrievalQuery)
+let d = try eg.encode(text: "Use FileManager.contentsOfDirectory(...)",
+                       task: .retrievalDocument)
+```
+
+See [`Gemma3EmbeddingGemma.swift`](https://github.com/john-rocky/CoreML-LLM/blob/main/Sources/CoreMLLLM/Gemma3EmbeddingGemma.swift)
+for task prefixes and dim list.
+<!-- swift-usage-end -->
+
+
+
 # EmbeddingGemma-300M for Apple CoreML (ANE-optimized)
 
 CoreML conversion of `google/embeddinggemma-300m` produced with the