Write agentic apps once, run with any LLM provider

Quickstart

Get up and running with AISuite in under 5 minutes. Learn how to install, configure, and make your first API call.

1. Installation

Install command:
pip install aisuite

You can also install with specific providers:

# Install with specific providers
pip install 'aisuite[anthropic]'
pip install 'aisuite[openai,anthropic]'
pip install 'aisuite[all]'  # Install all providers

2. Basic Usage

Initialize the client and make your first request:

example.py
import aisuite as ai

# Initialize client with provider configurations
client = ai.Client({
    "openai": {"api_key": "your-openai-key"},
    "anthropic": {"api_key": "your-anthropic-key"}
})

# Make a request to any provider
response = client.chat.completions.create(
    model="openai:gpt-4o",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Tell me a joke"}
    ]
)

print(response.choices[0].message.content)

3. Configuration Options

Using Environment Variables

The recommended way to manage API keys is through environment variables:

Environment variables
import aisuite as ai
import os

# Set API keys as environment variables
os.environ['OPENAI_API_KEY'] = 'your-openai-key'
os.environ['ANTHROPIC_API_KEY'] = 'your-anthropic-key'

# Client will automatically use environment variables
client = ai.Client()

# The provider will use the environment variable
response = client.chat.completions.create(
    model="anthropic:claude-3-5-sonnet-20240620",
    messages=[{"role": "user", "content": "Hello!"}]
)

4. Using Multiple Providers

One of AISuite's key features is the ability to use the same code with different providers:

Multi-provider example
import aisuite as ai

client = ai.Client()

models = ["openai:gpt-4o", "anthropic:claude-3", "google:gemini-pro"]

for model in models:
    response = client.chat.completions.create(
        model=model,
        messages=[{"role": "user", "content": "What is 2+2?"}],
        temperature=0.7
    )
    print(f"{model}: {response.choices[0].message.content}")

5. Model Format

AISuite uses a simple format for specifying models: provider:model

ProviderExample Models
OpenAIopenai:gpt-4o, openai:gpt-3.5-turbo
Anthropicanthropic:claude-3-5-sonnet-20240620
Googlegoogle:gemini-pro, google:gemini-pro-vision
Mistralmistral:mistral-large-latest
Groqgroq:llama3-8b-8192

Next Steps