Lunara is an end-to-end platform for building, deploying and operating AI voice agents on real telephony. They greet your callers, mirror their language, pull answers from your knowledge base, and trigger your APIs and CRM in real time.
Each client has a dedicated login URL — see Our partners below.
Built on Google Gemini Live (native audio) and Twilio. One platform for inbound, outbound, knowledge, tools, analytics and human handoff.
Unlimited agents with custom greeting, system prompt up to 20k chars, 8 native voices, temperature control.
The agent detects the caller's language and switches automatically — Russian, Romanian, English and more.
Point any Twilio number at your agent. Sub-second voice powered by Gemini Live native audio.
Place individual or bulk outbound calls. Use Twilio numbers or your own SIP trunk.
Upload PDFs, DOCX, MD or TXT per agent. We chunk, embed and inject the top matches into every call.
Let the agent call your APIs mid-call — order status, HubSpot / Salesforce / Bitrix lookups, leads, tickets.
Watch active calls in real time. Hand off to a human the moment the conversation needs a person.
Dual-channel recordings, transcripts, token usage, durations and per-agent dashboards.
Your numbers, your data, your prompts. A self-hosted alternative to Vapi / Retell / Bland.
One real-time pipeline from the caller's mouth to the AI's voice — and back.
Twilio PSTN or SIP routes the call to our voice webhook.
A WebSocket streams audio between the call and Gemini Live with your prompt, voice and knowledge.
Gemini answers in the caller's language. When it needs data, it calls your webhooks or CRM tools.
Recording, transcript, tokens and status are saved. Review, hand off, or feed outcomes back.
Each partner gets their own branded login URL on lunara.now. All entrances lead into the same shared Lunara workspace.
If you already have a partner account, head to Our partners to sign in.