Client

Object Hierarchy:

Description:

var connection = new Settings.Connection() {
    url = "http://127.0.0.1:11434/api"
};
var client = new Client(connection);
var chat = new Call.Chat(client, "llama3.2") {
    stream = true
};
chat.messages.add(new Message(chat, "user", "Hello!"));
var response = yield chat.send(chat.messages);

// Add tools to Chat before chatting
var read_file = new Tools.ReadFile(client);
var chat = new Call.Chat(client, "llama3.2");
chat.add_tool(read_file);

// Tools are automatically called when the model requests them
chat.messages.add(new Message(chat, "user", "Read README.md"));
var response = yield chat.send(chat.messages);

var chat = new Call.Chat(client, "llama3.2") {
    stream = true
};
chat.message_created.connect((msg, content) => {
    if (msg.is_content && msg.is_stream) {
        // Process incremental content
        print(content.chat_content);
    }
});

Content:

Properties:

• public Connection connection { get; set; }
  Connection configuration for this client.

Creation methods:

• public Client (Connection connection)

Methods:

• public async Chat chat (string model, string text, Options? options = null, Cancellable? cancellable = null) throws Error
  Legacy chat method for backward compatibility.
• public async Chat chat_execute (Chat call) throws Error
  Executes a pre-prepared Chat object.
• public async Embed embed (string model, string input, int dimensions = -1, bool truncate = false, Options? options = null, Cancellable? cancellable = null) throws Error
  Generates embeddings for the input text.
• public async Embed embed_array (string model, ArrayList<string> input_array, int dimensions = -1, bool truncate = false, Options? options = null, Cancellable? cancellable = null) throws Error
  Generates embeddings for an array of input texts.
• public async Generate generate (string model, string prompt, string system = "", Options? options = null, Cancellable? cancellable = null) throws Error
  Generates a response for the provided prompt.
• public async ArrayList<Model> models () throws Error
  Lists all available models (OpenAI-compatible).
• public async ArrayList<Model> ps () throws Error
  Lists currently running models on the Ollama server.
• public async Model show_model (string model_name) throws Error
  Gets detailed information about a specific model including capabilities.
• public async string version (Cancellable? cancellable = null) throws Error
  Gets the version of the Ollama server.

Inherited Members: