Documentation

Generation API

Generate safe, responsible AI content with built-in RAIL compliance and quality checks.

Overview

Generation endpoints use large language models to create content while ensuring safety and responsible AI standards. All generated content is automatically evaluated for RAIL compliance.

Response Type
Synchronous (direct)
Processing Time
2-8s depending on length
Available Models
GPT-4o, Gemini Pro

How it works: Submit your generation request → Content generated with safety filters → RAIL evaluation performed → Return safe content with scores

Endpoints

Common Use Cases

💬 Customer Support Chatbots

Build RAG-powered chatbots that provide accurate, grounded responses from your knowledge base.

/rag/chatContext-grounded generation

✍️ Content Improvement

Automatically improve AI-generated content that scores low on specific RAIL dimensions.

/repromptScore-based improvement

🛡️ Safe Content Pipelines

Generate content with automatic safety filtering. Only returns content meeting minimum RAIL score thresholds.

/protected_generateBuilt-in safety filters

📚 Knowledge Base Q&A

Answer user questions accurately using your own knowledge base with RAG architecture.

/rag/chatDocument-grounded answers

Ready to Generate Content?

Start building with responsible AI generation endpoints.