Wan 2.7
near-frontier quality at a third of the price
third on the with-audio leaderboard at a third of seedance's price, with four generation modes including instruction-based editing — the best value in video, full stop.
alibaba's tongyi lab shipped wan 2.7 in april with four modes: text-to-video, image-to-video, reference-to-video, and instruction-based video editing. that last one is rarer than it should be and genuinely useful.
it ranks third on artificial analysis for text-to-video with audio — above kling, above veo 3.1 — at a flat ten cents a second regardless of resolution. seedance sits two places higher and costs three to seven times more. if you are producing volume, this is where the maths lands.
the description above is ours, condensed from the ranking. pricing moves — check it on the vendor's own page before you rely on it.
- category
- ai video models
- pricing
- $0.10 / second, all resolutions
- website
- fal.ai
third on the with-audio leaderboard at a third of seedance's price, with four generation modes including instruction-based editing — the best value in video, full stop.
we researched this category against vendors' own pricing pages and licence files. that is where this line comes from — not from the vendor, and not from anything they paid for.