ByThen

ByThen uses a Key Visual generated at the start of your project to lock in the art style, color palette, and character direction. This acts as a visual anchor for every storyboard scene and video clip generated throughout the project, keeping all segments visually cohesive from the first scene to the last.

ByThen can generate videos ranging from 30 seconds up to 30 minutes in production. Final projects can then be edited and exported at up to a maximum duration of 60 minutes. All exports are rendered at 25 FPS with 720p resolution; 1080p and 4K support are listed as coming soon.

ByThen routes your scenes through a pool of seven AI video models: Seedance 1.5 Pro, Seedance 2.0, Kling 2.6, Kling 3 Omni, Kling 3.0 Motion Control, Veo 3.1, and Omnihuman. The platform selects the most appropriate model per scene based on the motion requirements and visual context of each clip.

Yes. ByThen features a native Model Context Protocol (MCP) server, which allows AI agents such as Claude (via Claude Desktop or Claude Code) to generate, edit, and orchestrate long-form video projects entirely through natural language commands, making video production a programmable capability within an AI workflow.

Both options are available. In the Content Wizard, you can either let ByThen generate a full script from a topic prompt, or switch to 'Use My Script' and paste your own finished script (or upload a .doc or .pdf file). ByThen will then build the Key Visual and Storyboard directly from your provided copy without modifying it.