Getting started
FAQ
Quick answers to the questions we hear most.
How much does it cost?
Ready-made solutions start from $500. Final pricing depends on scope — from a simple agent to a large custom build — and you'll get an exact quote after a quick chat.
Do you use my data to train models?
No. Your data is used to run your agent, not to train shared models. With self-hosting, your data never leaves your own servers at all.
Which models do you use?
We're model-agnostic — open models like Llama, Mistral, Qwen, or DeepSeek, or commercial APIs like GPT, Claude, and Gemini. We pick per task and budget. See Models we support.
Can it run on our own servers?
Yes — self-hosted via Docker, or on a managed API. It's your call, and we can migrate later without rebuilding the agent. See Self-hosted vs API.
What languages are supported?
Agents work across many languages; we tune them to the languages your customers actually use.
What happens when the agent can't handle something?
It hands off to a person with the full context of the conversation. You stay in control of where that line sits.
Is there support after launch?
Yes — we monitor performance, tune behaviour, and keep improving the agent as your needs change.