BETA · privacy LLMs & voice servers operational · GPU upgrade underway for faster responses · packages may still change Status & Roadmap →
Transparent pricing · No hidden fees

Smart AI assistants that actually respect your data

Five clear plans — from free exploration to full AI sovereignty. With private inference on our EU servers and the option to use your own LLM keys, models or speech server.

🇪🇺 EU servers GDPR & ISO 27001 Cancel monthly Start free, no credit card
Monthly Yearly 15% off

Discover

Free — Discover

€0 forever

No credit card required

  • 1 AI bot
  • 1 team members (seats)
  • 1 domain per bot
  • 500 credits/month
  • Voice chat: from Grow
  • Privacy-mode: from Grow
  • BYOK: from Grow
  • Promo footer present
  • Community
Start for free

Launch

For your first business AI assistant

29 /month

Monthly billing

  • 1 AI bot
  • 2 team members (seats)
  • 1 domain per bot
  • 5.000 credits/month
  • Voice chat: from Grow
  • Privacy-mode: from Grow
  • BYOK: from Grow
  • No "Powered by" footer
  • Email (48h)
Choose this plan

Scale

Enterprise-grade AI with BYOK, BYOLLM and BYOSpeech

249 /month

Monthly billing

  • 10 AI bots
  • 15 team members (seats)
  • 10 domains per bot
  • 100.000 credits/month
  • 300 min voice/month
  • Privacy-mode (our own GPUs)
  • BYOK — your own API key
  • BYOLLM + BYOSpeech
  • No "Powered by" footer
  • Email + chat (8h)
Choose this plan

Sovereign

Full AI sovereignty on our platform

699 /month

Monthly billing

  •  AI bots
  •  team members (seats)
  •  domains per bot
  • 500.000 credits/month
  • 1.200 min voice/month
  • Privacy-mode (our own GPUs)
  • BYOK — your own API key
  • BYOLLM + BYOSpeech
  • No "Powered by" footer
  • Dedicated + SLA
Talk to sales
100% air-gapped · Enterprise

Self-Host Enterprise

For healthcare, government, defence and pharma with real air-gap or compliance requirements. We install Einstein on your own hardware (or via our partner) — no outbound traffic, all data stays within your infrastructure.

Hardware lease, dedicated CSM, custom integrations and SLA up to 99.95% available. Price based on seats, bots, conversation volume and KB size.

From €18,000/year — priced on your environment
Book a sales call → Response within 1 business day

Full feature comparison

All plans side-by-side — credits, voice, privacy and BYO options.

Feature Discover Launch Grow Scale Sovereign
Price
Monthly price €0 €29 €249 €699
Yearly price (15% off) €0 €295,80 €2.539,80 €7.129,80
Capacity
AI bots 1110
Seats (portal users) 1215
Domains per bot 1110
Credits per month (cloud) ¹ 5005.000100.000500.000
KB ingest messages/month 501.00050.000
Voice (speech-to-text)
Included voice minutes 300 1.200
Voice overuse €/min €0,08 €0,06
Privacy & BYO
Private inference (EU GPU)
Air-gapped (100%) Self-Host Enterprise only — see section above.
BYOK (your Anthropic/OpenAI key)
BYOLLM (your LLM endpoint)
BYOSpeech (your STT/TTS)
Widget & branding
Widget templates 1 widget template3 widget templatesAll templates + custom CSSAll templates + white-label
Custom widget host Subdomain of your own domain via CNAME.
Promo footer removed No "Powered by InterAIP".
Support
Response time CommunityEmail (48h)Email + chat (8h)Dedicated + SLA

¹ Messages in privacy-mode consume 5 credits per message (because of our own GPU infrastructure). Messages via BYOK (your own API key) count as 0.3 credits. You see one counter on the dashboard — no separate pools.

∞ = unlimited within fair use. Conversation volumes > 500,000/year are handled under Self-Host Enterprise.

How do credits and voice work?

One counter, transparent rates. No surprises afterwards.

💰 AI credits — one counter, three modes

Every bot response consumes credits. The rate depends on where inference runs — everything is deducted from the same monthly bundle.

  • Cloud (Anthropic / OpenAI): 1 credit per message — standard, fast models.
  • Privacy-mode (our EU GPUs): 5 credits per message — your data never leaves our EU environment.
  • BYOK (your own API key): 0.3 credits per message — you pay the LLM directly, we only charge platform overhead.

On the dashboard you see one combined counter with a breakdown per mode.

🎤 Voice chat — speech-to-text included

Voice runs on our own Whisper server in the EU. Audio never leaves our infrastructure and is not used for training.

  • Grow: 60 minutes/month included, then €0.10/min
  • Scale: 300 minutes/month included, then €0.08/min
  • Sovereign: 1,200 minutes/month included, then €0.06/min
  • Discover & Launch: voice chat is disabled

Dutch and English speech recognition. No wholesale dependency on third parties.

How does Launch compare to popular alternatives?

Comparison based on each provider's closest entry plan.

What you get InterAIP.ai
Launch — €29
Crisp
Pro — €95
Watermelon
Premium — €106
Tidio
Starter — $59
Entry price/month €29 €95 €106 $59
Hosting in EU/NL Ja Frankrijk EU VS
Private inference on own GPU From Grow (€89) Nee Nee Nee
BYOK / own LLM key From Grow Nee Nee Nee
Voice chat (STT) From Grow (60 min/mo) Nee Nee Nee
White-label widget (no promo) Ja, vanaf Launch From Unlimited (€295) Add-on From €398/mo
Dutch interface + support Ja Partial Ja Nee

Competitor pricing was taken from their public pricing pages on 21 April 2026. We show the plan that's closest to Launch in features. Our Launch tier is deliberately accessible; from Grow we add features (privacy, voice, BYOK) none of these competitors offer in their base plan.

Frequently asked questions

Question not covered? Get in touch — we respond within 1 business day.

Privacy-mode routes your conversations to our own GPU infrastructure within the EU instead of to Anthropic or OpenAI. Your content never leaves our managed environment and is not used for model training. We call this private inference.

"100% air-gapped" (fully offline, on your own hardware) is only available in our Self-Host Enterprise variant. For most customers private inference is sufficient; for regulated sectors requiring air-gap we offer Self-Host.

Privacy messages run on our own EU GPUs rather than a commercial cloud API. That hardware is more expensive per inference — the 1:5 ratio reflects that cost fairly.

You keep one combined counter on the dashboard, so you don't have to juggle two bundles. Want a lower rate? Activate BYOK (your own API key) — messages then count as 0.3 credits.

With yearly billing you pay monthly price × 12 × 0.85 up front. Example: Grow is €89/month, or €907.80/year — you save €160.20 compared to monthly invoices.

Switching between yearly and monthly is possible any time from the portal. Upgrading to a higher plan within the year settles the difference pro-rata.

BYOK — Bring Your Own Key. You supply your own Anthropic, OpenAI or Google API key. Usage is billed to your account, not to platform credits. Available from Grow.

BYOLLM — Bring Your Own LLM. Configure an OpenAI-compatible endpoint per bot — a self-hosted Llama/Mistral, a fine-tune, or another cloud. Available from Scale.

BYOSpeech — Bring Your Own Speech. A custom Whisper/STT endpoint per tenant, for audio that must never leave your infrastructure. Available from Scale.

Sovereign (€699/mo) runs on our platform, with full route control via BYOK + BYOLLM + BYOSpeech. No on-prem hardware, but your own keys/endpoints/speech server. Fits customers with their own Azure/AWS LLM or a local inference stack.

Self-Host Enterprise (from €18,000/year) runs on customer-owned hardware on-premise. We install and support it, hardware leased via an InterIP partner or owned directly. Only needed for true air-gap (no outbound traffic allowed), such as healthcare, government, defence and pharma.

In short: Sovereign = the "logical top" of the tier ladder; Self-Host = a separate category for anyone who cannot have any cloud dependency.

Ready to try Einstein?

Start free with Discover, or upgrade straight to Launch. No credit card required to begin.

Create a free account →