About Yandex Cloud AI Studio

Yandex Cloud AI Studio is a platform for building AI agents and leveraging large generative models effectively.

Model Gallery offers a wide range of generative models for working with text and images. You can use them to:
Generate product descriptions, articles, news stories, newsletters, blog posts, advertising images, logos, and much more. To access the models running in the common instance, use AI Playground or integrate them into your apps with ML SDK and various APIs. If the model is missing from the common instance or is required for heavy workloads, you can deploy a dedicated instance right in the management console. For large amounts of homogenous data, you might want to use batch processing mode.

In addition to generative models, Model Gallery offers text classifiers and embedding models. If you are not satisfied with the response format of generative models or the behavior of classifiers and embeddings, you can fine-tune them.

The Agent Atelier development environment supports building text and voice AI agents that can use ready-made customizable tools. If ready-to-use tools are not enough, you can always build a custom MCP server in MCP Hub to enable the agent to use any external tools and APIs.

For those who prefer creating agents in a graphical editor, AI Studio offers a Yandex Workflows builder.

The AI Studio API is fully compatible with the OpenAI API: use the Responses API to create text agents, the Realtime API to create voice agents, and the Vector Store API to search through documents. The APIs support integration with such widely used frameworks as LangChain and LangGraph.

For information on the AI Studio restrictions, refer to Yandex AI Studio quotas and limits.