Wan 2.7 AI Video Generator

Wan 2.7 expands the text-to-video framing options beyond the basic landscape, vertical, and square set. Alongside 16:9, 9:16, and 1:1, it adds 4:3 and 3:4 for creators who need more editorial flexibility. 4:3 can feel natural for documentary-style shots, classic compositions, tutorials, and scenes where a wider vertical crop would lose important side detail. 3:4 works well for portrait-forward visuals, fashion content, product framing, and social creative that needs more vertical height without going fully 9:16. GenMix shows these ratios only where the provider supports them: text-to-video. For image-to-video, the shape comes from your uploaded first frame.

Many image-to-video workflows force users to write a long prompt even when the uploaded image already communicates most of the creative direction. Wan 2.7 supports optional prompt input for image-to-video, so you can keep the instruction light: a slow dolly-in, natural hair movement, floating particles, product rotation, soft breeze, cinematic lighting shift, or subtle handheld motion. When the source image is the key creative asset, this keeps the workflow fast and reduces over-specification. When you do need stronger direction, add a detailed prompt to guide action, timing, camera behavior, or atmosphere.

Wan 2.7 pricing is shown before you generate, with four simple combinations. A 720p 5-second clip costs 75 credits, 720p 10 seconds costs 150 credits, 1080p 5 seconds costs 110 credits, and 1080p 10 seconds costs 220 credits. The structure is intentionally linear by duration and separated by resolution, so creators can decide whether a clip is for draft iteration, social posting, client review, or final delivery. 720p is useful for exploring prompt ideas quickly, while 1080p is the better choice for sharper final clips and content that will be viewed on larger screens.