# Protify AI -- Open-Source Java AI SDK (Full Reference)

## Summary
Protify AI is an open-source Java SDK (Apache 2.0) that provides a unified API for 13+ AI providers including OpenAI, Anthropic, Gemini, Mistral, Groq, DeepSeek, and more. Built by Protify Consulting LLC, it has zero external dependencies, requires Java 11+, and supports single requests, multi-turn conversations, streaming, pipeline orchestration, tool use, MCP, declarative AI services, and structured output.

## Contact
- Email: jkuryla@protify.com
- Website: https://protify.ai
- GitHub: https://github.com/protifyconsulting/protifyai-java
- Company: Protify Consulting LLC (https://protify.com) — a technology consulting firm providing fractional CTO services, technology strategy, AI-enhanced application development, and software process leadership

## Features in Detail

### Single Requests
One-shot AI calls with text, images, and PDFs. Structured output deserializes JSON responses directly into Java objects. Simple builder pattern for constructing requests.

### Multi-Turn Conversations
Stateful dialogue with full message history. Pluggable persistence lets any service instance resume any conversation. Conversations maintain context across multiple exchanges.

### Pipeline Orchestration
Chain sequential, parallel, conditional, loop, and retry steps -- each targeting a different AI provider. Compose pipelines inside pipelines. Get per-step token stats. Supports complex multi-model workflows.

### Streaming
Real-time token-by-token output from any request or pipeline. Register listeners and get the full response when complete. Works with all providers.

### Tool Use & MCP
Define tools, detect tool calls, run manual or automatic tool loops. Connect to MCP servers via stdio or HTTP transport. Full function calling support across providers.

### Zero Dependencies
Uses only java.net.http and built-in JSON. Drop it into any Java project -- microservice, Android, CLI, legacy monolith. No framework lock-in.

### Declarative AI Services
Define annotated interfaces and let the SDK generate implementations via JDK proxies. Template variables, automatic return type mapping, streaming, and async -- zero boilerplate.

### Provider Agnostic
13 built-in providers -- OpenAI, Anthropic, Gemini, Mistral, Groq, DeepSeek, Together, Fireworks, xAI, Azure OpenAI, Vertex AI, AWS Bedrock, and Azure AI Foundry. Swap with a single config change. Pluggable interface for custom providers.

### Resilience Built In
Configurable retry policies with exponential backoff, jitter, status code filtering, and Retry-After header support. Production-ready error handling.

### Configuration Hierarchy
Four-level config cascade: base > client > pipeline > request. Override at any level with full control. Flexible and composable configuration.

## Supported Providers

1. OpenAI -- GPT-5 family
2. Anthropic -- Claude 4 family
3. Google Gemini -- Gemini Pro & Flash
4. Mistral -- Mistral Large & Small
5. Groq -- Ultra-fast inference
6. DeepSeek -- DeepSeek V3 & R1
7. Together -- Open-source models
8. Fireworks -- Optimized inference
9. xAI -- Grok models
10. Azure OpenAI -- Enterprise deployments
11. Google Vertex AI -- GCP managed
12. AWS Bedrock -- AWS managed
13. Azure AI Foundry -- Azure managed
14. Custom -- Any LLM API via pluggable interface

## Quick Start

### Gradle
```
implementation 'ai.protify:protifyai:0.1.9'
```

### Maven
```xml
<dependency>
    <groupId>ai.protify</groupId>
    <artifactId>protifyai</artifactId>
    <version>0.1.9</version>
</dependency>
```

### Basic Usage
```java
import ai.protify.ai.*;

public class App {
    public static void main(String[] args) {
        AIClient client = AIClient.builder()
                .model(AIModel.CLAUDE_SONNET_4_6)
                .build();

        AIResponse response = client.newRequest()
                .addInput("What is the capital of France?")
                .build()
                .execute();

        System.out.println(response.text());
    }
}
```

## Architecture
Your Application > Protify AI SDK (AIClient, AIRequest, AIConversation, AIPipeline, AITool, MCPClient > Config, Pipeline Engine, HTTP, JSON) > Providers (OpenAI, Anthropic, Gemini, Mistral, Groq, DeepSeek, Together, Fireworks, xAI, Azure OpenAI, Vertex AI, AWS Bedrock)

## Stats
- 0 external dependencies
- 4 interaction patterns
- 0 frameworks required
- Java 11+ minimum version
- 13+ providers supported

## Pages
- Home: https://protify.ai/