Skip to main content

Token Factory product overview

For your information

The product is in public preview.

Token Factory is a service for unified access to generative AI models deployed on Selectel infrastructure in Russian data centers. Requests are processed without cross-border data transfer.

Token Factory allows you to work with available models through a single interface with pay-per-token billing, centrally manage access keys, and obtain consolidated reporting and analytics on their usage.

If you need unified access to models deployed on third-party provider infrastructure (OpenAI, Anthropic, and others), use AI Router.

To work with Token Factory, projects are supported.

Tasks solved

  • working with models without cross-border data transfer;
  • connecting models via a single public OpenAI-compatible API key;
  • paying for models from a single balance;
  • monitoring consumption through detailed real-time analytics.

How it works

Token Factory acts as a universal gateway between your application and models deployed on Selectel infrastructure. The service sends requests to models via an OpenAI-compatible API; for details, see the official OpenAI Developers documentation. You do not need to configure integrations with each model separately — simply use the Token Factory API key in requests.

The contents of requests and model responses are not stored and are not used to fine-tune models. Token Factory collects only technical statistics to analyze product usage without prompt contents.

How to work with Token Factory

To get started, use the Token Factory: Quick start instructions.

After gaining access to Token Factory, you can:

Available models

You can view the current list of models, their specifications, and token prices in the Control panel: in the top menu, click Products → Token Factory → tab Model catalog.

Areas of responsibility

Selectel provides

  • access to models via a public OpenAI-compatible API;
  • the Token Factory monitoring system in the Control panel;
  • consumption analytics;
  • ability to manage API keys;
  • technical support.

Selectel is not responsible for

  • the content and use of model responses;
  • the operability of requests made with API keys;
  • processing personal data within the service;
  • integrating models into your projects;
  • the business logic of model operations in your projects.

Protecting API keys from compromise is your responsibility.

Cost

Token Factory is billed using the pay-as-you-go model. Funds are debited from the balance every hour for tokens consumed during the previous hour; for details, see Token Factory billing model and pricing.

Limits and restrictions

In one project, you can create up to ten API keys.

Token Factory does not support:

  • access to models deployed on third-party provider infrastructure;
  • the ability to train models;
  • uploading custom models to the catalog;
  • deploying models from the catalog on-premises, in A-DC, on dedicated servers, and in Compliant Cloud.