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.

flux-video-upscale from replicate-flux works with text: it accepts text queries and documents as input and returns text. The context is 4095 tokens, so the model is convenient for short tasks: analyzing a document fragment, writing a description, rewriting, or structuring a small text.

In one response, the model outputs up to 4096 tokens — this is several pages of text. The total context window is 4095 tokens, so it is better to divide long requests and voluminous responses into parts and process the material sequentially, in small fragments.

A detailed reasoning mode is not noted in our data, but that does not mean it does not exist. If the task requires logic, ask the model to write out the solution step-by-step directly in the prompt — in practice, this noticeably improves the quality of the response.

Support for function calling and strict JSON output is not noted in our data; the easiest way to check is with a test query. If the response structure is critical, describe the required format in the prompt. In BotHub, all models are available via a unified OpenAI-compatible API, so you can switch to another model without rewriting the integration.

Besides text, the model accepts documents — you can attach a file and ask for a summary, find specific points, or rephrase a fragment. Acceptance of images, audio, and video is not noted in our data. Keep in mind the 4095-token window: it is better to upload large files in parts.

We have no data on the quality of work with the Russian language, so we will not make any claims — evaluate it on your own task: send a couple of typical queries and compare the result with another model in the same window. Access to BotHub works from Russia without a VPN or foreign card, payment is in rubles.