| What it does | Build, launch and optimize your own customer-facing agent
| AI Operating System for building agentic AI solutions and automating conversations
|
|---|
| Stage | Not given
| Established
|
|---|
| Founded | Not given
| Not given
|
|---|
| Based in | DE
| IT
|
|---|
| Pricing model | Freemium
| Freemium
|
|---|
| Pricing | Free: €0 forever, 50 evidence answers a month, 1 verified website. Starter: €39/month, 750 evidence answers a month, 1 verified website, 7-day trial available. Business: €149/month, 3,000 evidence answers a month, 3 verified websites, 7-day trial available. Scale: from €449/month, 12,000 evidence an
| Free trial: 14 days with all premium features. Pro: €149/month (€124/month billed yearly), includes 3 users, 20 AI Agents, 5M AI Tokens per month, 3,000 conversations per month. Business: €499/month (€416/month billed yearly), includes 5 users, 50 AI Agents, 10M AI Tokens per month, 5,000 conversati
|
|---|
| Key features | - Build from your own content: import web pages, PDFs, Word documents, text files and Q&A
- Source attribution: every answer shows the documents it came from
- Integrations with Zapier, n8n, Pipedrive, HubSpot, Slack, Microsoft Teams, Google Sheets
- Multiple AI models including Claude, OpenAI, DeepSeek and Mistral
- Handoffs to live agents: direct visitors to contact forms or booking pages
- Private testing before going live
- Monitor unanswered questions and add approved answers to improve the agent
- Transparent pricing with no hidden overage charges
- Monthly reset with no rollover and 80% usage notification
| - Visual flow builder for AI agent design without coding
- Native knowledge base with RAG for accurate AI responses
- Multi-channel deployment across web, WhatsApp, voice, SMS, email and Facebook Messenger
- Human-in-the-loop with seamless escalation to human agents
- Prompt chaining for complex multi-step automation workflows
- MCP (Model Context Protocol) integration with external tools and APIs
- Multiple specialized AI agents with automatic routing
- Self-learning capabilities that improve from human interactions
- On-premises deployment with local LLMs and vector stores
- AI copilot for human agents with real-time assistance
|
|---|
| What makes it different | - Only uses your approved content, not the open internet
- Says when it doesn't know instead of guessing
- Every answer shows its sources with citations
- Free to test, pay only for working answers
- No credit card required for free plan
| - Open-source platform with full control over source code and deployments
- No-code visual builder requiring zero programming knowledge
- Agentic architecture supporting task execution beyond text generation via MCP
- Self-learning agents that automatically improve from real conversations
- Deployment flexibility with cloud, on-premises, and local LLM options
|
|---|
| Free plan | Not given
| Not given
|
|---|
| Open source | Not given
| Not given
|
|---|
| Platforms | Not given
| Not given
|
|---|
| Public API | Not given
| Not given
|
| Community votes | 0 | 0 |
| DR (Domain Rating by Ahrefs) | 41 ● no change since last check | 62 ● no change since last check |
| Trust Flow (Majestic) | 13 | 11 |
| Citation Flow (Majestic) | 26 | 32 |
| Referring domains (Majestic) | 14 | 1044 |