You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.

GigaChat-Max is a text model from GigaChat: it writes and edits texts, summarizes documents, answers questions, and helps with code and analytics. The 128,000-token context allows you to upload a long report, correspondence, or technical documentation and work with them in their entirety. In BotHub, it is available without a VPN, and payment is possible with Russian cards.

According to our data, the output limit is up to 128,000 tokens per response, the same as the context window. In practice, this is enough for a voluminous article, a detailed analysis of a document, or a large fragment of code. If the response turns out to be longer, just ask it to continue.

A separate reasoning mode before answering is not noted in our data for this model, so we will not promise it. However, you can ask GigaChat-Max to break down a task step-by-step directly in the prompt — step-by-step instructions usually significantly improve the quality of analyzing complex questions.

In our data, there is no mention of function calling or structured output for this model, but this does not mean the capability is absent. It is convenient to set the required format in the prompt: describe the schema and ask it to return only JSON, and verify the result on your end.

In our data, GigaChat-Max is noted for working with text; the acceptance of other file types is not marked. Focus on text scenarios: insert transcripts, exports, and documents as text. If you need images, audio, or video, specialized models are available nearby in BotHub — you can switch in the same window.

GigaChat is a domestic development, and working with Russian-language texts is a basic scenario for this line of models: it handles cases, names, business, and conversational styles confidently. The best way to check is to provide your own real material and compare the response with other models in the same BotHub window.