Your AI Chatbot.
€79 once.Forever yours.

No. Run bash setup.sh — the interactive wizard configures your domain, generates secrets, installs Docker if needed, and starts everything. Setup time: under 5 minutes. Detailed docs included.

OpenAI (GPT-5.5, GPT-4o, o3), Anthropic (Claude Opus 4.7, Sonnet 4.6, Haiku 4.5), Google Gemini (3.1 Pro, 2.5 Flash), OpenRouter (any vendor/model — GPT, Claude, Llama, Mistral, and more), and a Custom/Compatible provider that works with any OpenAI-compatible endpoint — including Groq, Together AI, and self-hosted LLMs like Ollama. You can switch providers at any time without touching your data.

Yes. The server runs on your infrastructure — your VPS, your cloud, your hardware. Conversation data, uploaded knowledge files, and captured leads never leave your server. The only external calls are to your chosen AI provider's API (which you control).

Yes. Remove all "AI Chat Agent" branding, set your own logo, avatar, color scheme, and domain. The widget and admin panel are fully customizable. Under the Regular License you can host and operate chatbots for your paying clients under your own brand — agency and managed-service use is explicitly allowed. The only thing the Regular License does not cover is redistributing the source code itself; for that you'd need an Extended License. Details in our Terms of Use.

From a single admin panel you can create as many bots as you need. Each bot is fully independent: it has its own system prompt, knowledge base, AI provider selection, widget appearance, analytics, leads inbox, and notification settings. Switch between bots using the sidebar switcher — 2 clicks. Each bot gets a unique embed code (data-bot-id). Ideal for agencies managing multiple clients.

Yes. The Custom provider option accepts any OpenAI-compatible endpoint. Point it at your Ollama, LM Studio, or any locally-served model. Set the base URL, API key (if required), and model name — done. 100% offline operation is possible.

Yes. Your purchase includes free updates for the lifetime of the product. New features, security patches, and bug fixes are delivered through your account.

It's actively developed — the current release is v1.5.1, with features and fixes shipping regularly (recent additions: OpenRouter as a 5th AI provider, RAG grounding so the bot won't hallucinate off-topic, EN/RU widget localization, and visitor-identity + UTM passthrough). And because the complete source code ships with your license, you're never locked out: in the worst case you own a full, working codebase you can run and modify yourself — forever.

You can — and for some use cases you should. But the popular free options solve different problems: Chatwoot is a live-chat/helpdesk inbox, Typebot is a flow builder, LibreChat is a ChatGPT-style UI. None of them ship a RAG knowledge base, multi-bot management, a fully white-label embeddable widget, and lead capture as one product you deploy in 5 minutes. AI Chat Agent does — and you get the complete source code plus lifetime updates for €79 once, instead of assembling and maintaining a stack of separate tools yourself.

No. When a bot has a knowledge base attached, retrieval uses a similarity threshold — if your documents don't actually cover the question, the bot tells the visitor it doesn't have that information and offers a human handoff, rather than guessing. You can also take over any conversation yourself in real time with Operator Live Reply.

Docker and Docker Compose, 1GB RAM minimum (2GB recommended), and an API key from any supported AI provider. Any Linux VPS works — DigitalOcean, Hetzner, AWS, you name it. Cold start under 90 seconds. Handles 10 concurrent users with under 2s latency on minimal specs.