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.

The model works with text, images, and documents: it answers questions, writes and edits content, analyzes code, and extracts data from files and screenshots. A 65,536-token context allows you to keep long correspondence, a large article, or an entire voluminous document in a single dialogue.

Up to 32,000 tokens are available per response — this is a very long text, enough for an article, detailed documentation, or a large code snippet. If there is even more material, break the task into parts and continue in the same dialogue without losing context.

There is no specific note about reasoning before answering in our data, so we will not promise a "thinking" mode. In practice, a simple trick helps: ask it to break down the task step-by-step and show the logic — for complex questions, the answers turn out more accurate.

There is no note about function calling or strict JSON mode in our data, so we cannot guarantee them. Usually, an explicit instruction helps: describe the required schema and ask it to return only JSON. You can connect the model to your code via the unified OpenAI-compatible BotHub API.

Images and documents are confirmed: upload screenshots, photos, PDFs, and ask it to analyze the content. There are no notes about audio and video in our data, so you shouldn't count on them — for audio and video, BotHub has separate models in the same window.

There are no separate measurements for Russian in our data, so we won't invent an assessment — it's easier to test it with your own task. In BotHub, it's inexpensive: pay-as-you-go for tokens, no subscription, no VPN or foreign card required, and other models are available in the same window for comparison.