Frequently Asked Questions¶

The Cavuer AI agents are powered by GPT and Claude, which are among the most powerful language models in the world. You can freely switch between the available models at any time. We are also planning to add models from providers such as Gemini and DeepSeek in the near future.

Yes, our AI agents support more than 60 languages.

Yes, the agents can read information from images and “understand” voice messages.

Depending on the configured delay, the agent can correctly interpret such requests. The longer the delay, the longer the agent will wait before responding, giving the user time to complete their question if it is sent in several messages. You can adjust the delay in your personal account in the agent settings.

You can configure instructions for the agent to handle spam based on your needs.

The number of agents and channels depends on your subscription level.

To upload information, you need to add the necessary documentation to the knowledge base. Learn more in the Knowledge Base article.

You can browse your knowledge base, add new articles, and edit previously uploaded documents. Learn more in the Knowledge Base article.

Agent responses are charged in credits. On average, one agent response costs 1 credit. You can find more details on the Pricing page.

Upon registration, you will have access to the Free plan, and complimentary credits will be added to your account balance. The Free plan includes access to various language models (GPT-4.1, GPT-4.1 mini, Claude Haiku family) and features, allowing you to upload documentation, create agents, and test our service in action.

The Free, Optimal, and Premium plans are available in your account dashboard. These plans differ in their monthly credit allowance, the number of available agents and channels, limits on knowledge base article uploads, and access to various language models and features. You can find a detailed comparison of all plans on the Pricing page.