Every AI model in one chat. Pay only for tokens

Riser is a single interface for working with different AI models and ready-made AI personas. You can use models from different providers without purchasing a separate subscription for each service.

There are no plans or monthly fees. You top up your balance and pay only for the tokens actually used.

Riser uses pay-as-you-go billing. You top up your balance and the cost of each request is deducted after it has been processed.

The price depends on the selected model and the number of input and output tokens. There are no subscriptions, monthly fees or automatic recurring charges.

Tokens are units used by AI models to process text. Your message, the relevant conversation history and the model's answer are all converted into tokens.

Input and output tokens may have different prices. A longer conversation or answer usually requires more tokens and therefore costs more.

An estimated cost is shown in the chat before you send a request. The forecast takes into account the selected model, your message and the conversation context available at that moment.

The estimate is not the final charge because the exact length of the model's answer is not known in advance. After processing, you are charged only for the tokens actually used.

The exact number of output tokens is only known after the model generates its answer. The final charge may therefore be slightly lower or higher than the forecast.

Conversation history also affects the price: in a long chat, more context may be sent to the model with each new request.

An AI persona is a ready-made preset that combines a master prompt, task-specific instructions, scenarios and a preferred AI model.

A persona helps you start working without configuring everything manually. It determines the role, communication style and approach to the task, while requests are still billed by the tokens used by its model.

A model is the AI engine that processes your request. When you select a model directly, you decide how to formulate the task and can use the available project or chat settings.

A persona is a configured assistant built on top of a model. It already includes a master prompt, role, working scenarios and recommended behavior. Both options use the same token-based billing.

Choose a persona based on your task: writing, learning, marketing, analytics, career development, image generation or another area.

The persona card shows its description, category and preferred model. You can start with a ready-made persona or create your own preset with custom instructions and a selected model.

Yes. You can create a personal persona by writing a master prompt, defining its role and instructions, and selecting a suitable AI model.

This is useful for recurring tasks where you want to keep the same tone, rules, workflow or output format without repeating them in every request.

Riser provides access to text, image and other AI models from multiple providers, including OpenAI, Anthropic, Google, DeepSeek, xAI, Qwen and others.

The catalog is updated as models are added, replaced or archived. You can view the currently available models, their capabilities and prices on the AI models page.

Yes. Riser supports connecting your own API keys for compatible AI providers. This allows requests to be processed through your provider account while you continue using the Riser interface, projects and chat history.

Availability and billing conditions may depend on the selected provider. API keys are stored in encrypted form.

You can invite colleagues to a workspace, use shared or private chats and manage work with AI models in one interface.

A shared balance makes it possible to pay for team usage from one account. Access to chats and projects depends on workspace permissions.

If a request was not processed because of a technical error, or you believe the amount was charged incorrectly, please contact Riser support. Include the approximate time of the request, the model or persona used and a description of the issue.

If the model generated an answer but it did not meet your expectations, try clarifying the request, adding context or selecting another model. AI-generated answers may contain errors and should be checked before being used for important decisions.

Requests are transmitted to the selected AI provider so that the model can process them and generate an answer. Different providers may apply different data processing policies.

Messages and connected API keys are stored and transmitted using encryption. Avoid sending passwords, payment details, trade secrets or unnecessary personal data. For more information, see the Privacy Policy.

Riser is continuously adding new models, features and improvements. You can follow announcements, participate in discussions and share feedback in our Telegram community: @riser_chat.