diff --git a/.swiftpm/xcode/xcshareddata/xcschemes/LangChain.xcscheme b/.swiftpm/xcode/xcshareddata/xcschemes/LangChain.xcscheme
index b5ca24a..ab04876 100644
--- a/.swiftpm/xcode/xcshareddata/xcschemes/LangChain.xcscheme
+++ b/.swiftpm/xcode/xcshareddata/xcschemes/LangChain.xcscheme
@@ -1,6 +1,6 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <Scheme
-   LastUpgradeVersion = "1500"
+   LastUpgradeVersion = "1540"
    version = "1.7">
    <BuildAction
       parallelizeBuildables = "YES"
diff --git a/.swiftpm/xcode/xcshareddata/xcschemes/LangChainTests.xcscheme b/.swiftpm/xcode/xcshareddata/xcschemes/LangChainTests.xcscheme
index be06304..7cedf8e 100644
--- a/.swiftpm/xcode/xcshareddata/xcschemes/LangChainTests.xcscheme
+++ b/.swiftpm/xcode/xcshareddata/xcschemes/LangChainTests.xcscheme
@@ -1,6 +1,6 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <Scheme
-   LastUpgradeVersion = "1500"
+   LastUpgradeVersion = "1540"
    version = "1.7">
    <BuildAction
       parallelizeBuildables = "YES"
diff --git a/.swiftpm/xcode/xcshareddata/xcschemes/langchain-swift-Package.xcscheme b/.swiftpm/xcode/xcshareddata/xcschemes/langchain-swift-Package.xcscheme
index 837e804..b272af8 100644
--- a/.swiftpm/xcode/xcshareddata/xcschemes/langchain-swift-Package.xcscheme
+++ b/.swiftpm/xcode/xcshareddata/xcschemes/langchain-swift-Package.xcscheme
@@ -1,6 +1,6 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <Scheme
-   LastUpgradeVersion = "1500"
+   LastUpgradeVersion = "1540"
    version = "1.7">
    <BuildAction
       parallelizeBuildables = "YES"
diff --git a/Sources/LangChain/llms/ChatOllama.swift b/Sources/LangChain/llms/ChatOllama.swift
index badcaef..4c7ed99 100644
--- a/Sources/LangChain/llms/ChatOllama.swift
+++ b/Sources/LangChain/llms/ChatOllama.swift
@@ -15,7 +15,7 @@ public class ChatOllama: Ollama {
     ///
     /// This array contains the chat history
     /// of the conversation so far.
-    var history = [ChatGLMMessage]()
+    public var history = [ChatGLMMessage]()
 
     /// Create a new Ollama chat instance.
     ///
@@ -60,26 +60,26 @@ public extension ChatOllama {
     /// This is a streaming endpoint, so there can be a series of responses.
     /// Streaming can be disabled using "stream": false.
     struct ChatRequest: Codable, Sendable {
-        let model: String
-        let options: [String: String]?
-        let format: String
-        let stream: Bool
-        let messages: [ChatGLMMessage]
+        public let model: String
+        public let options: [String: String]?
+        public let format: String
+        public let stream: Bool
+        public let messages: [ChatGLMMessage]
     }
     /// Ollama response to a `ChatRequest`.
     ///
     /// This response object includes the next message in a chat conversation.
     /// The final response object will include statistics and additional data from the request.
     struct ChatResponse: Codable, Sendable {
-        let message: ChatGLMMessage
-        let model: String
-        let done: Bool
-        let totalDuration: Int?
-        let loadDuration: Int?
-        let promptEvalDuration: Int?
-        let evalDuration: Int?
-        let promptEvalCount: Int?
-        let evalCount: Int?
+        public let message: ChatGLMMessage
+        public let model: String
+        public let done: Bool
+        public let totalDuration: Int?
+        public let loadDuration: Int?
+        public let promptEvalDuration: Int?
+        public let evalDuration: Int?
+        public let promptEvalCount: Int?
+        public let evalCount: Int?
 
         /// Return the message content.
         public var content: String { message.content }
diff --git a/Sources/LangChain/llms/Ollama.swift b/Sources/LangChain/llms/Ollama.swift
index 07b2cff..955325c 100644
--- a/Sources/LangChain/llms/Ollama.swift
+++ b/Sources/LangChain/llms/Ollama.swift
@@ -14,10 +14,10 @@ import OpenAIKit
 /// This class is a wrapper around the Ollama API.
 public class Ollama: LLM {
 
-    let baseURL: String
-    let model: String
-    let options: [String: String]?
-    let requestTimeout: Int
+    public let baseURL: String
+    public let model: String
+    public let options: [String: String]?
+    public let requestTimeout: Int
 
     /// Images encoded as base64 strings.
     public var images: [String]?
@@ -100,6 +100,7 @@ public class Ollama: LLM {
             return LLMResult()
         }
         let llmResponse = try JSONDecoder().decode(GenerateResponse.self, from: data)
+        context = llmResponse.context
         return LLMResult(llm_output: llmResponse.response)
     }
 
@@ -163,17 +164,17 @@ public extension Ollama {
     /// This message is sent to the `generate` API
     /// endpoint to generate a response.
     struct GenerateRequest: Codable {
-        let model: String
-        let prompt: String
-        let images: [String]?
-        let system: String?
-        let template: String?
-        let context: [Int]?
-        let options: [String: String]?
-        let keepAlive: String?
-        let format: String?
-        let raw: Bool?
-        let stream: Bool?
+        public let model: String
+        public let prompt: String
+        public let images: [String]?
+        public let system: String?
+        public let template: String?
+        public let context: [Int]?
+        public let options: [String: String]?
+        public let keepAlive: String?
+        public let format: String?
+        public let raw: Bool?
+        public let stream: Bool?
 
         enum CodingKeys: String, CodingKey {
             case model
@@ -188,23 +189,55 @@ public extension Ollama {
             case raw
             case stream
         }
+
+        /// Create a new `GenerateRequest`.
+        ///
+        /// This initializer creates a new `GenerateRequest`.
+        ///
+        /// - Parameters:
+        ///  - model: The model to use for generation.
+        ///  - prompt: The prompt to use for generation.
+        ///  - images: An array of images to use for generation.
+        ///  - system: The system prompt to use for generation.
+        ///  - template: The template to use for generation.
+        ///  - context: The context to use for generation.
+        ///  - options: The options to use for generation.
+        ///  - keepAlive: The keep-alive time for the generation.
+        ///  - format: The format to use for the response.
+        ///  - raw: Whether to return the raw response.
+        ///  - stream: Whether to stream the response.
+        /// - Returns: A new `GenerateRequest`.
+        @inlinable
+        public init(model: String, prompt: String, images: [String]? = nil, system: String? = nil, template: String? = nil, context: [Int]? = nil, options: [String: String]? = nil, keepAlive: String? = nil, format: String? = nil, raw: Bool? = nil, stream: Bool? = nil) {
+            self.model = model
+            self.prompt = prompt
+            self.images = images
+            self.system = system
+            self.template = template
+            self.context = context
+            self.options = options
+            self.keepAlive = keepAlive
+            self.format = format
+            self.raw = raw
+            self.stream = stream
+        }
     }
     /// Ollama generation response.
     ///
     /// This response object contains the response
     /// generated by the Ollama `generate` API endpoint.
     struct GenerateResponse: Codable {
-        let response: String
-        let createdAt: String
-        let model: String
-        let done: Bool
-        let context: [Int]?
-        let totalDuration: Int?
-        let loadDuration: Int?
-        let promptEvalDuration: Int?
-        let evalDuration: Int?
-        let promptEvalCount: Int?
-        let evalCount: Int?
+        public let response: String
+        public let createdAt: String
+        public let model: String
+        public let done: Bool
+        public let context: [Int]?
+        public let totalDuration: Int?
+        public let loadDuration: Int?
+        public let promptEvalDuration: Int?
+        public let evalDuration: Int?
+        public let promptEvalCount: Int?
+        public let evalCount: Int?
 
         enum CodingKeys: String, CodingKey {
             case response
@@ -219,6 +252,38 @@ public extension Ollama {
             case promptEvalCount = "prompt_eval_count"
             case evalCount = "eval_count"
         }
+
+        /// Create a new `GenerateResponse`.
+        ///
+        /// This initializer creates a new `GenerateResponse`.
+        ///
+        /// - Parameters:
+        ///  - response: The generated response string.
+        ///  - createdAt: The creation time of the response.
+        ///  - model: The model used for generation.
+        ///  - done: Whether generation is complete.
+        ///  - context: The context used for generation.
+        ///  - totalDuration: The total duration of the generation.
+        ///  - loadDuration: The load duration of the generation.
+        ///  - promptEvalDuration: The duration of the prompt evaluation.
+        ///  - evalDuration: The duration of the evaluation.
+        ///  - promptEvalCount: The prompt evaluation count.
+        ///  - evalCount: The evaluation count.
+        /// - Returns: A new `GenerateResponse`.
+        @inlinable
+        public init(response: String, createdAt: String, model: String, done: Bool, context: [Int]? = nil, totalDuration: Int? = nil, loadDuration: Int? = nil, promptEvalDuration: Int? = nil, evalDuration: Int? = nil, promptEvalCount: Int? = nil, evalCount: Int? = nil) {
+            self.response = response
+            self.createdAt = createdAt
+            self.model = model
+            self.done = done
+            self.context = context
+            self.totalDuration = totalDuration
+            self.loadDuration = loadDuration
+            self.promptEvalDuration = promptEvalDuration
+            self.evalDuration = evalDuration
+            self.promptEvalCount = promptEvalCount
+            self.evalCount = evalCount
+        }
     }
     /// Generate the next message in a chat with a provided model.
     ///
@@ -270,11 +335,16 @@ public extension Ollama {
     /// This struct represents a single Ollama model
     /// that is available via the API.
     struct Model: Codable {
-        let name: String
-        let modifiedAt: String
-        let digest: String
-        let size: Int
-        let details: ModelDetails
+        /// Ollama model details.
+        public let name: String
+        /// The last modification date of the model.
+        public let modifiedAt: String
+        /// The digest of the model.
+        public let digest: String
+        /// The size of the model.
+        public let size: Int
+        /// The details of the model.
+        public let details: ModelDetails
 
         /// JSON coding keys for the `Model` struct.
         enum CodingKeys: String, CodingKey {
@@ -289,11 +359,16 @@ public extension Ollama {
     ///
     /// This struct represents the details of an Ollama model.
     struct ModelDetails: Codable {
-        let format: String
-        let family: String
-        let families: [String]?
-        let parameterSize: String
-        let quantizationLevel: String
+        /// The format of the model.
+        public let format: String
+        /// The family of the model.
+        public let family: String
+        /// The families of the model.
+        public let families: [String]?
+        /// The size of the model parameters.
+        public let parameterSize: String
+        /// The quantization level of the model.
+        public let quantizationLevel: String
 
         /// JSON coding keys for the `ModelDetails` struct.
         enum CodingKeys: String, CodingKey {
diff --git a/Sources/LangChain/utilities/chatglm/ChatGLMAPIWrapper.swift b/Sources/LangChain/utilities/chatglm/ChatGLMAPIWrapper.swift
index 263b5c7..5e09681 100644
--- a/Sources/LangChain/utilities/chatglm/ChatGLMAPIWrapper.swift
+++ b/Sources/LangChain/utilities/chatglm/ChatGLMAPIWrapper.swift
@@ -9,9 +9,15 @@ import AsyncHTTPClient
 import Foundation
 import NIOPosix
 
-struct ChatGLMMessage: Codable {
-    let role: String
-    let content: String
+/// A ChatGLM message.
+///
+/// This structure represents a message in a chat.
+public struct ChatGLMMessage: Codable, Sendable {
+    /// The role of the entity sending the message,
+    /// such as "user", "assistant", or "developer".
+    public let role: String
+    /// The content of the message.
+    public let content: String
 }
 struct ChatGLMPayload: Codable {
     let prompt: [ChatGLMMessage]