Adds support for the embeddings endpoint

Signed-off-by: ActuallyTaylor <[email protected]>

Sources/SLlama/Extensions/URLRequest+Request.swift

@@ -13,6 +13,8 @@ extension URLRequest {
 
         httpMethod = request.method.name
         httpBody = try request.method.httpBody
+
+        addValue("application/json", forHTTPHeaderField: "Content-Type")
     }
 }
 

Sources/SLlama/LlamaRequests.swift

@@ -11,4 +11,8 @@ public enum LlamaRequests {
     public static func completion(_ parameters: CompletionSettings) -> Request<CompletionResult> {
         return Request(path: "/completion", method: .post(.body(model: parameters)))
     }
+
+    public static func embedding(_ parameters: EmbeddingSettings) -> Request<EmbeddingResults> {
+        return Request(path: "/embedding", method: .post(.body(model: parameters)))
+    }
 }

Sources/SLlama/Models/EmbeddingResults.swift

@@ -0,0 +1,12 @@
+//
+//  File.swift
+//  
+//
+//  Created by Taylor Lineman on 11/19/23.
+//
+
+import Foundation
+
+public struct EmbeddingResults: Codable {
+    public let embedding: [Double]
+}

Sources/SLlama/Models/EmbeddingSettings.swift

@@ -0,0 +1,13 @@
+//
+//  File.swift
+//  
+//
+//  Created by Taylor Lineman on 11/19/23.
+//
+
+import Foundation
+
+public struct EmbeddingSettings: Codable {
+    /// Text to process
+    let content: String
+}

Tests/SLlamaTests/SLlamaTests.swift

@@ -14,6 +14,16 @@ final class SLlamaTests: XCTestCase {
 
     """
 
+    func testEmbeddings() async throws {
+        let client = Client(baseURLString: "http://127.0.0.1:8080")
+
+        let settings: EmbeddingSettings = .init(content: "This is some contents to embed")
+
+        let request = LlamaRequests.embedding(settings)
+
+        _ = try await client.run(request)
+    }
+
     func testCompletionEndpoint() async throws {
         let client = Client(baseURLString: "http://127.0.0.1:8080")
 
@@ -22,15 +32,14 @@ final class SLlamaTests: XCTestCase {
 
         let request = LlamaRequests.completion(settings)
 
-        let response = try await client.run(request)
-
-        print(response.content)
+        _ = try await client.run(request)
     }
 
     func testCompletionStreaming() async throws {
-        let client = Client(baseURLString: "http://127.0.0.1:8080")
+        let client = Client(baseURLString: "http://127.0.0.1:24445")
 
         let preparedPrompt = PromptProcessor.prepareTemplate(template: template, systemPrompt: "A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the human's questions.", userPrompt: "Hello Assistant")
+
         let settings: CompletionSettings = .init(prompt: preparedPrompt, temperature: 0.7, n_predict: 256, stream: true)
 
         let request = LlamaRequests.completion(settings)
@@ -50,7 +59,7 @@ final class SLlamaTests: XCTestCase {
 extension SLlamaTests: ClientStreamDelegate {
     func didReceiveModel<Model>(model: Model) where Model : Codable {
         guard let model = model as? CompletionResult else { return }
-        print(model.content)
+        print(model.content, terminator: "")
     }
 
     func didFinish(error: Error?) {