Seedance 1 Pro Fast is silent I2V

Seedance 1 Pro Fast is silent image-to-video: 50 credits for 5s and 100 for 10s, 1080p, audio false. The still is required. The soundtrack is not in the row. If the brief is a talking pack, a multilingual line, or a pour that has to sound like a pour, this is the wrong Seedance.

The still matters because the category is image-to-video. This page is the flag: mute on purpose, 50 credits, two lengths, 1080p. It is also not Seedance 2.0. 2.0 is 44 credits for text-to-video and 97 for reference-to-video, native audio, 1080p, 4–15 seconds. Fast 1 Pro does not inherit that microphone.

The card: category image-to-video only, image required, audio false, durations 5s and 10s, 1080p, 50 credits for 5s and 100 for 10s, not marked discounted. Description: accelerated image-to-video generation with high visual quality. Acceleration is speed. It is not a reason to skip the still, and it is not a reason to expect speech.

Audio false is the job

audio: false is the useful fact. There is no native dialogue, no room tone you can quote in a client review, no Foley the model will invent because you wrote “glass clink.” You will get motion on the still, quiet.

That is a real job. Mute autoplay still ships. A product orbit that will get a desk VO later still ships. A draft that exists to see whether this pack can move without melting the type still ships. What does not ship is a presenter line you buried in the prompt. The flag does not flip because you asked nicely.

Do not prompt “add voiceover” and rerun Fast. Change rows. Seedance 2.0 is the talking cinematic Seedance at 44 credits. Wan 2.6 Image to Video is 20 credits, image-to-video, native audio true, 1080p, 5 / 10 / 15 seconds. Veo 3.1 is 64 credits, text-to-video, native audio, 4K, 4 / 6 / 8 seconds. None of those are this SKU. Fast is the mute 50-credit I2V.

Native audio is a catalog field, not a vibe. False means you plan the mix. Captions for silent autoplay is the typesetting if the feed is mute-first. Fast does not typeset. It moves the frame.

When Fast is the pick:

• The still is locked and you need 5 or 10 seconds of motion at 1080p.
• The voice will be a creator, a desk agent, or a TTS row laid after.
• The placement is mute-first and a burned-in line is the audio.
• You will throw the take away if the label crawls. Fifty credits is how you find that out on this family, not how you buy a soundtrack.

When Fast is the wrong pick:

• Anyone has to speak in-model.
• You have no still. This row will not start from a paragraph.
• You needed 4, 6, 8, 15, or 30 seconds. The menu is 5s and 10s.
• You needed 4K. The ceiling here is 1080p.

5s or 10s at 1080p, 50 or 100 credits

supports_durations is 5s and 10s. There is no 6s, no 8s, no 4s test, no 15s hold. A 5s generate is the draft length. 10s is the long hold on this card. Cutting a 10s take down in an editor is fine. Paying for 10s because you wanted 6s and the slider felt adjacent is not. The catalog does not sell 6s on this slug.

Resolution is 1080p. Do not rerun Fast hoping it becomes 4K. VEO First Last Frame is the 4K I2V sibling in another family: 64 credits, native audio, 4 / 6 / 8 seconds. That is a different bill and a different microphone. Fast is 1080p mute.

Fifty credits is the catalog sticker for 5 seconds. The card is not marked discounted. Do not average it with Seedance 2.0’s 44 and call Fast “almost the same price so it must talk.” 50 versus 44 is mute I2V versus native-audio T2V, not a coupon.

Seedance 1.5 Pro Text to Video is 4 credits, text-to-video, native audio, 1080p, 4–12 seconds. That is the cheap talking Seedance sentence. It is not this row. Fast is expensive relative to 1.5 Pro because it is a different category: still in, mute motion out, accelerated. If you wanted a cheap talking sentence, you are on 1.5 Pro, not on 50-credit Fast.

Draft versus ship on this SKU:

1. Approve the still at 100%. Soft type will not sharpen in video.
2. Run 5s. If the label crawls, fix the still, not the prompt.
3. Run 10s only when the hold is the brief.
4. Lay sound somewhere else, or leave the file mute and burn a caption.

Do not run 10s first “to see the motion” on an unapproved pack. You will spend the long sticker on a draft still. Do not run 5s and then be surprised it is quiet. You picked the quiet row.

When Fast is the wrong family

Seedance 2.5 is 34 credits (was 67), text-to-video, native audio, 4–30 seconds, 720p ceiling. Fast 1 Pro is not a 30-second row and not a talking row. If length is the job, Grok Imagine Video is 24 credits, native audio, 720p, 6 / 10 / 15 / 20 / 30 seconds. If 4K dialogue in 8 seconds is the job, Veo 3.1 is 64. If a stack of stills must survive a new scene, Seedance 2.0 reference-to-video at 97 is the door, not Fast.

Kling 3 Turbo Text to Video is also silent, 22 credits, 3–15 seconds, 1080p, no still required. Mute is not one row. Turbo has no still slot. Fast has no sentence-only slot. Pick by whether the plate exists.

Wan 2.6 Image to Video Flash is 20 credits, image-to-video, audio false, 1080p, 5 / 10 / 15 seconds. That is the cheap mute I2V draft in another family. Fast at 50 is not the cheap mute draft. It is ByteDance’s accelerated 1 Pro I2V at 1080p with a 5s/10s menu. Pay 50 when that motion is the file. Do not pay 50 because you wanted Flash’s 20-credit mute and clicked the wrong Seedance.

Sora 2 leftovers that were mute product pushes can land here if you have the still and 1080p is enough. Sora’s API shuts 24 September 2026 (OpenAI Help Centre, September 2026). Sora 2 T2V was 32 credits, native audio, 720p. Fast is not a Sora replacement for talking 720p. It is a mute 1080p I2V sticker. Map the flag first. The Sora API sunset migration plan is the longer export note. This page is the Seedance mute door.

After the mute take

A silent 50-credit clip is not a finished ad. Burn a caption. Lay a VO if the brief needs a voice. Crop. The AI caption generator is the typesetting. Cartesia Sonic 3.5 is 4 credits if you need TTS. Suno Sounds v5.5 is 2 credits if you need a bed. Native audio on another generate row is the other way to get a soundtrack. Pick one. Doing both without a plan is how you stack a VO on a VO.

Article 50 still applies to photoreal people in the still. Fast being mute does not exempt a generated face. Label at first exposure for EU audiences since 2 August 2026. The Article 50 disclosure template is the copy. YouTube’s 27 May 2026 labelling pass overlays photoreal AI on Shorts. Labels do not add a soundtrack.

The AI video generator is the launcher. Treating a still as UGC does not change the flag. Handheld-looking product motion still needs a mix. Fast will not invent one.

Fifty credits is the motion bill. The mix is a second line on the slate. Write that line before you press generate, or you will discover the silence at upload.

Run the mute I2V in Versely.

FAQ

Does Seedance 1 Pro Fast have native audio?

No. The published card is audio false. There is no in-model speech on this slug. Change rows if the take has to talk.

Why pay 50 when Seedance 1.5 Pro T2V is 4 credits?

1.5 Pro is a talking text-to-video sentence. Fast is mute image-to-video from a still at 1080p for 5s or 10s. Same brand, different category, different flag.

Can I ship Fast as the final file?

Yes, when mute plus captions (or a VO you lay) is the ship. Do not ship Fast to a brief that required native dialogue.