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.

muse-spark-1.1 works with text and documents, reads images, reasons before answering, and calls functions. It is suitable for parsing large reports and code, extracting data from PDFs and scans, and preparing structured responses for your services. The context is 1,048,576 tokens, so you don't need to split materials.

The output limit is 943,718 tokens per response, which is enough for a voluminous report, documentation, or a series of chapters. In practice, it is more convenient to request results in parts: this makes it easier to check quality as you work and manage token consumption, which does not expire.

Yes, the reasoning mode is confirmed: the model first builds a solution path and then provides the result. This helps with tasks involving multi-step logic, code analysis, calculations, and document reconciliation. Note that reasoning also consumes tokens, so complex requests should be formulated precisely.

Yes, both function calling and structured JSON output are supported, making the model easy to integrate into agents and pipelines. In BotHub, this is available via a unified OpenAI-compatible API: you describe the schema once, and then change the model without rewriting the integration.

You can input images and documents, and files are read natively without separate recognition. Audio and video are not marked in our data, so for voiceovers or analyzing recordings, choose a specialized model and switch to it in the same BotHub window.

We have no data on languages, so we cannot promise accuracy in Russian—please check with your typical requests. However, access from Russia works without a VPN or foreign card, payment is made with Russian cards, and there are other models in the interface for comparison.