Website | Documentation

---

The definitive, open-source Swift framework for interfacing with generative AI.

1. Open your Swift project in Xcode.
2. Go to File -> Add Package Dependency.
3. In the search bar, enter this URL.
4. Choose the version you'd like to install.
5. Click Add Package.

+ import AI

Initialize an instance of LLMRequestHandling with an API provider of your choice. Here's an example:

import AI
import OpenAI

let llm: any LLMRequestHandling = OpenAI.APIClient(apiKey: "sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx")

You can now use llm as an interface to an LLM as provided by the underlying provider.

In the snippet earlier, we initialized an instance of LLMRequestHandling with OpenAI as the underlying provider.

OpenAI offers a number of chat models, like GPT-3.5 and GPT-4.

A chat model is a language model that can be given a set of messages and asked to generate a response that follows in turn.

You can use the LLMRequestHandling.complete(_:model:) function to generate a chat completion for a specific model of your choice. For example:

/// ...

let llm: any LLMRequestHandling = OpenAI.APIClient(apiKey: "sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx")

let messages: [AbstractLLM.ChatMessage] = [
    AbstractLLM.ChatMessage(
        role: .system,
        body: "You are an extremely intelligent assistant."
    ),
    AbstractLLM.ChatMessage(
        role: .user,
        body: "Sup?"
    )
]

let result = try await llm.complete(
    messages,
    model: OpenAI.Model.chat(.gpt_4)
)

print(result) // "Hello! How can I assist you today?"

In this example we constructed an ordered array of chat messages, and used our llm instance to generate a completion using GPT-4.

• OpenAI
• Anthropic
• Mistral
• Ollama
• Perplexity
• Groq

• similarity-search-kit
• Tiktoken

This package is licensed under the MIT License.