relai handles your customer conversations on WhatsApp, Instagram, Messenger, and your own website — powered by the AI of your choice, with your team always in control.
Built lean for SMBs and sales teams across Southeast Asia. Powerful enough to scale.
Switch between Claude, OpenAI GPT-4o, and Google Gemini without touching a single config file. Your keys, your models, your costs — fully under your control.
The AI checks your Google Calendar and books appointments mid-conversation — zero manual steps, zero double bookings.
Add team members with admin or agent roles. Everyone sees the same live threads, can jump in at any time, and hands off without missing a beat.
Trigger words escalate to a human instantly. Draft mode lets you review every reply before it goes out. Toggle per conversation, any time.
Live WebSocket feed of every conversation. Watch the bot work, jump in at any moment, see platform-wide stats, and toggle the AI per thread.
API keys and tokens are AES-256-GCM encrypted. Masked in the UI, locked in the DB. Your secrets never leave your infrastructure.
One script tag adds an AI chat bubble to any website. Visitors get instant answers; you see every thread in the same dashboard as your social channels.
A message arrives on WhatsApp, Instagram DMs, or Facebook Messenger. relai receives it in milliseconds via webhook or local bridge.
Your chosen AI (Claude, GPT-4o, or Gemini) gets the full conversation history, your persona prompt, and optional calendar tools. It crafts a reply that sounds like your team.
In live mode the reply goes out with a human-like delay. In draft mode it's routed to your team for approval. You decide, per conversation.
Every message streams live to the dashboard over WebSocket. Agents can take over, add notes, and hand back to the bot — all from one screen.
Connect all your messaging channels — social and web — to a single AI-powered workspace.
QR-linked bridge or Meta Cloud API — works either way
Reply to DMs automatically with context-aware AI responses
Handle Messenger conversations at scale, effortlessly
Embed live AI chat on any website with a single script tag
AI replies cost tokens. Move the sliders to see how your chat volume maps to monthly usage — and which plan fits.
relai uses prompt caching — your system prompt and conversation history are cached after the first turn, so repeated reads cost ~10% of normal input price. Without caching, each exchange would run ~2.5× higher.
On the BYOK plan you pay your AI provider directly — token volume affects your provider bill, not your relai subscription.
We're doubling down on automation so your team spends time closing, not chasing.
Full relai dashboard in your pocket. Get push notifications for escalations, approve AI drafts, and jump into any conversation — from anywhere.
Every relai setup is tailored to your business. Schedule a free live demo and we'll walk you through exactly how it works for your team.
relai is currently in private beta. Fill in your details and we'll reach out when we're ready to onboard you.
No spam. We'll only email you when it's your turn.