API Reference Jido AI v0.5.2

View Source

Modules

High-level API for accessing AI provider keys.

A low-level action that provides direct access to Instructor's chat completion functionality. Supports most Instructor options and integrates with Jido's Model and Prompt structures.

A low-level action that provides direct access to Langchain's chat completion functionality. Supports multiple providers through Langchain adapters and integrates with Jido's Model and Prompt structures.

Action module for generating vector embeddings using OpenAI Ex.

Action module for generating images using OpenAI Ex.

Helper module for handling tool calling with OpenAiEx. Provides functions to convert Jido.Action to OpenAI tool format and handle tool responses.

General purpose AI agent powered by Jido

The Jido.AI Application module.

Custom error type for Jido.AI operations.

A GenServer that manages environment variables and application configuration.

Defines the behavior that any provider adapter must implement.

A module that provides struct-based prompt generation.

Represents a single message item within a prompt.

This module is used to split a string into chunks by the number of tokens, while accounting for other data that might be going with it to the API endpoint with the limited token count.

Enables defining a prompt template and delaying the final building of it until a later time when input values are substituted in.

Protocol for data structures that can be converted into prompts for LLMs.

Adapter for the Anthropic AI provider.

Adapter for the Cloudflare AI provider.

Adapter for the Google Gemini API provider.

Common helper functions for AI provider adapters.

Adapter for the OpenAI AI provider.

Adapter for the OpenRouter AI provider.

General purpose AI skill powered by Jido

A simple tokenizer for AI models that provides basic encode/decode functionality. This is a placeholder implementation that should be replaced with actual tokenization logic.

A demo module showcasing how to use Jido.AI with Instructor for structured prompting.

Example demonstrating how to use Google's Gemini models with Jido AI.

Mix Tasks

Fetches and caches models from AI providers.