Model Select or Create - IronLabs Docs

IronlabsAI recommends the best-suited LLM to respond by analyzing an array of messages and a list of available LLMs, letting you handle the LLM calls however you like.

When to use `model_select`

Use model_select if you want to integrate IronlabsAI into an existing project. The lightweight ironlabsai package minimizes dependency conflicts and works smoothly with your existing code for making LLM requests.

selected_models = client.chat.completions.model_select(
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain the golden ratio."}
    ],
  	models=['openai/gpt-4o', 'anthropic/claude-3-5-sonnet-20240620']
)

## Response output
# selected_models = 'openai/gpt-4o'

const result = await client.modelSelect({
  messages: [{ content: 'Explain the golden ratio.', role: 'user' }],
  models: ['openai/gpt-4o', 'anthropic/claude-3-5-sonnet-20240620'],
});

When to use `create`

Use create if you’re starting a new project & want to skip writing extra code for LLM calls. Most examples here use create, but you can switch to model_select anytime.

completion_response, selected_models = client.completions.create(
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain the golden ratio."}
    ],
  	models=['openai/gpt-4o', 'anthropic/claude-3-5-sonnet-20240620']
)

const result = await client.create({
  messages: [{ content: 'Explain the golden ratio.', role: 'user' }],
  models: ['openai/gpt-4o', 'anthropic/claude-3-5-sonnet-20240620'],
});