Yes. reAPI.ai runs every Seedance tier below fal, by 44% to 77%. Both platforms bill per second of output. Figures are USD per second at 720p, text-to-video, 16:9, audio on:
reAPI.ai is a pay-as-you-go multimodal API. One key covers Seedance alongside image, video and chat models, one credit is 0.001 USD, credits never expire, and a failed generation refunds in full automatically.
Why those per-second numbers carry a caveat
fal bills Seedance by token rather than by second: tokens = height × width × duration × 24 / 1024. The fal column is its own published per-second figure for the common 16:9 case, but its docs call the token formula authoritative, so a 9:16 clip has a different frame area and a different effective rate. reAPI.ai bills per second of output directly, so the published rate holds at any aspect ratio.
Both platforms bill a reference video on top of the output rather than instead of it. An eight-second reference feeding a five-second clip bills thirteen seconds on either side, which is where per-second comparisons usually go wrong.
Full 480p to 4K tiers sit on each reAPI.ai model page above. Rates read from both platforms' public model pages on 20 August 2026.
发表于