UAITO SDK

The UAITO SDK provides a powerful command-line interface (CLI) to interact with a variety of large language models (LLMs) from different providers. Whether you're working with cutting-edge models from OpenAI and Anthropic or running local models with ONNX, this SDK simplifies the process and provides a clean, streamlined experience.

CLI Usage

The CLI is designed to be straightforward and easy to use, with a focus on providing a clean, readable output. By default, it operates in a streaming mode, allowing you to see the response as it's generated, and it suppresses all internal logging to keep the output tidy.

Run Command

The primary command is run, which allows you to send a message to the specified model and receive a response.

npx @uaito/sdk run "What is the capital of France?" --provider OpenAI --model gpt-5-nano --apiKey openAIKey

Arguments

• <message>: (Required) The message you want to send to the agent, enclosed in quotes.

Options

• --provider, -p: (Required) The LLM provider to use.
  • choices: OpenAI, Anthropic, Local
• --model, -m: (Required) The specific model to use from the chosen provider.
  • OpenAI Models: gpt-5-nano, gpt-5-mini
  • Anthropic Models: claude-4-sonnet
  • Local Models: JANO, LUCY, QWEN3
• --apiKey: The API key for the selected provider. This is only required for OpenAI and Anthropic. Alternatively, you can set the OPENAI_API_KEY or ANTHROPIC_API_KEY environment variables.

Examples

Using a Remote Model (OpenAI)

To get a response from OpenAI's gpt-5-nano model, you would run:

npx @uaito/sdk run "What is the capital of France?" --provider OpenAI --model gpt-5-nano --apiKey YOUR_OPENAI_API_KEY

Using a Local Model

If you have a local ONNX model, you can run it without needing an API key:

npx @uaito/sdk run "Translate 'hello' to Spanish" --provider Local --model JANO

The CLI will stream the response directly to your console, providing a clean and readable output.