Meet Iris AI
|
Turn your WordPress site into an intelligent assistant. Iris AI indexes your content, finds the right answer, and replies with sources — so visitors get real answers, not generic chat.
What Iris does
Answers based on your content, branded UI, traffic-safe rate limits, and full usage visibility.
Answers from Your Site
Your assistant responds using your actual pages, posts, and docs — with cited sources — instead of guessing.
Semantic Retrieval (RAG)
High-quality embeddings and similarity search so it understands meaning, not just keywords.
Drop-In Chatbox
Add an AI chat widget to any page. Visitors can ask questions and get grounded answers in real time.
Brand-Ready
Control colors, welcome message, tone, behavior and more — so it feels native to your site, not bolted on.
Usage Log & Insights
See what people asked, which content was cited, and how many tokens each answer used.
Rate Limits & Abuse Protection
Protect your budget. Iris can throttle or block excessive requests so strangers can't drain your allowance.
How usage & billing works
Iris is token-metered. You choose who pays for the AI calls: either you (with your own key) or us (with prepaid tokens). The assistant works the same either way.
Option 1: Bring Your Own API Key
Connect your own OpenAI-compatible key directly in WordPress. Iris will call that model for answers. We don’t bill you, because you’re billed by your provider. You get full Iris behavior — semantic retrieval, cited answers, chatbox on your site, usage log — without buying credits from us.
- Full features: semantic RAG answers with sources, visitor chatbox, analytics.
- You control which model you use (including advanced models).
- Your key never leaves your site.
Option 2: Prepaid Tokens (Managed by Iris)
Don’t want to paste an API key into WordPress? You can let Iris host the model calls for you. You get a token balance, and each answer (including embeddings/indexing) deducts a small amount. You also get free starter tokens when you create your account in our web portal.
- No direct OpenAI setup — we handle the requests for you.
- See detailed usage logs in the dashboard: timestamp, question, which content was cited, and how many tokens were spent.
- Built-in rate limits and burst protection so random visitors can’t drain everything at once.
- When tokens run out, Iris can throttle or pause answers instead of surprise-charging you.
Your site already has answers.
Iris makes them available in a friendly chat — with citations. Drop it on your site and watch support questions disappear.