# Powerful AI model capabilities

## Strong multimodal models, one creation workspace

U9 AI brings video, image, music, audio and available text model APIs into a workspace with clear prices, request records and spending analysis. Choose by the output your project needs and test the exact configuration before integrating.

| Capability | What to check in the live configuration |
| --- | --- |
| Text-to-video | Supported duration, resolution, aspect ratio and prompt controls |
| Image-to-video | Reference image requirements, motion prompt, output settings |
| Image generation | Supported inputs, output dimensions and generation mode |
| Music and audio | Model-specific inputs, output format and billing unit |
| Text and coding | Available model ID, Chat Completions / Responses / Messages support, tools and streaming |

## Test the capability that matters to your project

Open the [live models and pricing](/en/models), select a supported configuration and use the [creation workspace](/en/app#create). Start small, compare output quality, processing time and settled spending. The [model testing guide](/en/docs/model-testing) explains the procedure and test-credit inquiries.

Reference images, Markdown prompt uploads and advanced controls improve control where supported; they do not guarantee identical results or universal benchmark leadership. Catalog presence does not establish every protocol or feature. See [clear answers](/en/docs/faq) and [API compatibility](/en/docs/api).

## From a test to an API integration

Download the [tool integration guides](/en/docs/agent-files), choose an available model and verify the full request flow. Use private tickets for technical help, model access and test-credit questions. Tests that generate content are billed at the displayed configuration price unless support explicitly confirms a credit arrangement.