AIML

This component lets you create AI-powered chatbots within Nappai. It uses advanced AI language models to generate text responses based on your input. Think of it as a smart text generator that you can customize.

Relationship with AIML API

This component connects to the AIML API (located at https://api.aimlapi.com by default) to access and use powerful AI language models for generating text. You’ll need a valid AIML API credential to use this component.

Inputs

Max Tokens: This sets the maximum length of the AI’s response. A higher number means longer responses, but it can also take longer to generate. You can set it to 0 for unlimited tokens (though this is not recommended for performance reasons). The maximum value is 128000.
Model Kwargs: This allows for advanced customization of the AI model’s behavior using additional parameters. Unless you need very specific adjustments, you can leave this blank.
Model Name: Choose the specific AI model you want to use from a list provided by Nappai. The default option is usually a good starting point.
AIML API Base: This is the web address of the AIML API. Usually, you don’t need to change this, but it’s here if you need to connect to a different AIML API server.
Temperature: This controls how creative or random the AI’s responses are. A lower value (like 0.1, the default) produces more focused and predictable responses. A higher value (closer to 1) makes the responses more creative and unpredictable.
Seed: This setting helps ensure you get the same response each time you use the same input and settings. This is useful for testing and debugging.
Credential: This is your access key to the AIML API. Make sure you keep this secure!

Outputs

This component doesn’t have explicit outputs in the way some others do. Instead, it generates text based on your input and settings. This text can then be used by other components in your Nappai workflow, such as displaying the response to a user or using it as input for another AI process.

Usage Example

Imagine you want to build a chatbot that answers questions about your company’s products. You would configure the AIML component with:

Model Name: A suitable model for question answering.
Max Tokens: A value that allows for sufficiently long answers.
Temperature: A low value (e.g., 0.2) for precise answers.
Credential: Your AIML API key.

Then, you would feed user questions into the AIML component as input. The generated text (the chatbot’s answer) would be the output, which you can then display to the user.

Templates

[This section will be populated with links to relevant templates once they are available.]

Explain SQL results: Use this to get human-readable explanations of complex database query results. You could use the AIML component to make this explanation even more conversational.
Natural Language to SQL: This component converts natural language requests into SQL queries. The AIML component could then be used to explain the results of those queries.
[Add links to other relevant components as needed, with brief descriptions.]

Tips and Best Practices

Start with the default settings for the Model Name and Temperature. Adjust them only if you need more creative or specific responses.
Experiment with different values for Max Tokens to find the optimal balance between response length and generation time.
Always keep your AIML API Credential secure and do not share it publicly.

Security Considerations

Protect your AIML API Credential. Do not hardcode it directly into your Nappai workflow; use Nappai’s secure credential management features. Treat your credential like a password.