AI Models

Vozmi currently uses the following third-party AI models and access providers to deliver user-facing generation features. Model availability may change as providers update their services; this page will be updated when the production configuration changes.

• Music generation: Suno V4, accessed through the Kie.ai API.
• Image generation and storyboards: GPT Image 2 (gpt_image_2), accessed through the Aireiter API.
• Video generation: Seedance 2.0 Fast Reference-to-Video (seedance-2.0-fast-reference-to-video), accessed through the Evolink API.

The music-video workflow also uses supporting models for planning, transcription, review, and vocal analysis. These models help prepare a generation but are not presented as the final image or video generator.

• Music-video planning: GPT-5.6 Luna, connected through an OpenAI-compatible API endpoint.
• Plan review: DeepSeek V4 Flash, connected through an Anthropic-compatible API endpoint.
• Audio transcription: OpenAI Whisper Large V3 Turbo, accessed through OpenRouter.
• Vocal analysis: Xiaomi MiMo V2.5, accessed through OpenRouter.

To fulfill a generation request, Vozmi sends the necessary prompt, media, and technical settings to the applicable provider. Provider processing is subject to its own service terms and safeguards. Vozmi does not represent that it owns these third-party models. See our Privacy Policy for more information about data processing.

For questions about model access, licensing, or the configuration currently used by Vozmi, contact us.