What

Soutine AI is a browser-based generative media studio designed for creators, marketers, and designers seeking to produce visual assets without specialized local software. The platform consolidates text-to-image, image-to-image, text-to-video, and image-to-video generation by hosting multiple third-party and proprietary models, including Qwen Image 2.1, GPT Image 2.5, and Nano Banana variants. It positions itself as a unified creation hub where users can conceptualize static imagery, perform targeted edits, and extend stills into motion within a continuous workspace.

Features

Multi-Modal Generation: Supports text-to-image, image-to-image, text-to-video, and image-to-video pipelines to handle static design and motion rendering from a single interface.
Identity-Preserving Editing: Employs models like GPT Image 2.5 to alter specific elements—such as outfits, backgrounds, or styling—while maintaining consistent character identity, pose, and environmental context.
AI Video Upscaling: Enhances input video clips up to 4K resolution to restore fine visual details for higher-definition display.
Integrated Creative Workspace: Stores assets, parameters, and generated media in a shared history, allowing users to carry still images directly into video generation or upscaling workflows.

Helpful Tips

Check Commercial Licensing: Free tier assets are public, watermarked, and restricted from commercial use; commercial rights and watermark-free downloads require a paid subscription or a purchased credit pack.
Monitor Credit Expiration: Subscription credits expire at the end of each billing cycle without rollover, whereas separate one-time credit packs do not expire and suit irregular generation needs.
Verify Reference Formats: Ensure uploaded reference media uses standard PNG, JPEG, or WebP formats to maintain compatibility across image-to-image and image-to-video models.

OpenClaw Skills

The provided documentation contains no evidence of a native integration between Soutine AI and OpenClaw. In a hypothetical implementation, an OpenClaw autonomous agent could ingest creative briefs, invoke Soutine's generation interface to render multi-model image or video variants, and retrieve the final outputs for automated publishing workflows. If implemented, such an agent-driven pipeline could reduce manual production time for digital marketing and content teams by automating repetitive drafting, styling, and upscaling tasks.