Simple, transparent pricing
No seat fees. No surprise overage invoices. Pick a plan that fits your usage and scale up when you need to.
Pricing plans
For small teams testing AI support for the first time.
- 1 tenant / 1 bot configuration
- Web widget channel only
- PDF, DOCX, TXT ingestion
- 1,000 AI conversations per month
- Claude Haiku 4.5 LLM
- Basic analytics dashboard
- Email support
For growing teams that need multi-channel delivery and live operators.
- 10 tenants / unlimited bot configs
- Web widget + WhatsApp + Telegram
- All document formats — up to 50 MB
- 10,000 AI conversations per month
- Claude Haiku 4.5 LLM
- Human escalation + appointment booking
- Sentiment analysis per turn
- Full analytics + OpenTelemetry traces
- Bundled Grafana dashboards
- 2FA (TOTP) + encrypted secrets
- Multi-role users (admin / operator)
- Stripe billing integration
- Priority support
Full data sovereignty, compliance, and white-label rights.
- Unlimited tenants & bot configs
- Self-hosted Docker Compose deploy
- Ollama on-prem LLM (fully air-gapped)
- Ollama mxbai-embed-large embeddings
- Unlimited AI conversations
- White-label reseller rights
- All Growth features included
- Custom SSO / SAML
- Dedicated deployment support
- SLA with uptime guarantee
- Priority engineering access
Feature comparison
| Feature | Starter | Growth | Enterprise |
|---|---|---|---|
| Channels | |||
| Web widget (JS embed) | ✓ | ✓ | ✓ |
| WhatsApp (Meta Cloud API) | — | ✓ | ✓ |
| Telegram (aiogram) | — | ✓ | ✓ |
| RAG Pipeline | |||
| Voyage AI embeddings (voyage-3) | ✓ | ✓ | ✓ |
| Ollama on-prem embeddings | — | — | ✓ |
| pgvector HNSW index | ✓ | ✓ | ✓ |
| Document formats (PDF, DOCX, TXT) | ✓ | ✓ | ✓ |
| Document formats (MD, HTML, CSV) | — | ✓ | ✓ |
| File size limit | 10 MB | 50 MB | Unlimited |
| LLM | |||
| Claude Haiku 4.5 (default) | ✓ | ✓ | ✓ |
| Ollama on-prem LLM (phi4-mini, llama3.2…) | — | — | ✓ |
| Operators & Escalation | |||
| Confidence-scored escalation | — | ✓ | ✓ |
| Live operator WebSocket console | — | ✓ | ✓ |
| Appointment booking in widget | — | ✓ | ✓ |
| Sentiment analysis per turn | — | ✓ | ✓ |
| Analytics & Observability | |||
| Basic analytics dashboard | ✓ | ✓ | ✓ |
| OpenTelemetry tracing (OTel) | — | ✓ | ✓ |
| Bundled Grafana dashboards | — | ✓ | ✓ |
| Per-turn bot_traces table | — | ✓ | ✓ |
| Admin & Multi-Tenancy | |||
| Tenants | 1 | 10 | Unlimited |
| Bot configurations per tenant | 1 | Unlimited | Unlimited |
| Per-tenant personality / system prompt | ✓ | ✓ | ✓ |
| White-label reseller rights | — | — | ✓ |
| Multi-role users (superadmin / admin / operator) | — | ✓ | ✓ |
| Security | |||
| Argon2 password hashing | ✓ | ✓ | ✓ |
| JWT RS256 tokens | ✓ | ✓ | ✓ |
| 2FA (TOTP) via QR code | — | ✓ | ✓ |
| Encrypted secrets at rest | ✓ | ✓ | ✓ |
| Rate limiting + account lockout | ✓ | ✓ | ✓ |
| Self-host / zero third-party exposure | — | — | ✓ |
| Billing & Support | |||
| Stripe payment integration | ✓ | ✓ | ✓ |
| Crypto / manual payment providers | — | — | ✓ |
| Support | Priority | Dedicated + SLA | |
Pricing questions
A conversation is counted once per unique session — regardless of how many messages are exchanged within that session. A visitor who sends 10 messages in one sitting counts as one conversation.
Yes — add-on conversation packs are available on all plans. Alternatively, upgrading to the next tier or switching to Enterprise with unlimited conversations is often more cost-effective for high-volume use cases.
Self-hosted deployment is available on the Enterprise plan with custom pricing. Contact our sales team to discuss your infrastructure requirements, conversation volumes, and tenant count.
Annual billing is available with a discount equivalent to two months free. Reach out via the contact form or book a demo to discuss annual terms for your team.