Wan 2.7 Text to Video: pick the row, not the brand (2s, 3s, 4s, 1080p, 15cr)
Wan 2.7 Text to Video is one catalog generate. Use it because this job matches this row.
Wan 2.7 Text to Video is one catalog generate. Use it because this job matches this row.
Wan 2.7 Text to Video is Wan's high-fidelity text-to-video generate with strong motion consistency, optional custom audio input, and intelligent prompt rewriting. Durations: 2s through 15s. Resolution: 1080p. Credits: 15. No image required. Audio flag: empty. The AI video generator is where you run it. "Wan" is the family on the Wan roster. This line is the 2-second floor.
Two seconds is a real length
The ladder is 2s, 3s, 4s, 5s, 6s, 7s, 8s, 9s, 10s, 11s, 12s, 13s, 14s, 15s. Most text-to-video rows in this catalog start later. This one starts at two. A sting, a logo hit, a cutaway that only needs a breath — that is the 2s/3s/4s the title is for. A 15-second wander is also on the row. It is not the default personality of the model.
1080p ceiling. Qualities: 720p and 1080p. Aspects: 16:9, 9:16, 1:1, 4:3, 3:4. Pick the length first, then the ratio. A 2-second 9:16 sting is a different object from a 15-second 16:9 scene. Same row. Different job. Say which one you are buying.
Requires image: false. If the brief is a locked plate, you want image-to-video, including other Wan 2.7 lines, not this text generate.
Prompt rewriting is not a license to shrug
The description names intelligent prompt rewriting. Prompt expansion will fill gaps. Gaps it fills are not always your gaps. Write the shot: subject, action, camera, light. Let rewrite catch grammar, not invent a city you did not want.
Motion consistency is the other half of the description. One action per clip. A 4-second clip with two stunts is how consistency breaks. Use the 2-second floor for the cutaway instead of stuffing it into the master take.
Optional custom audio input is a file you bring
Audio on the row is not marked true. The description says optional custom audio input. That is the opposite of "the model will score this for you." If you have a track, this line can take it. If you do not, do not expect a native bed just because other Wan rows have audio: true.
15 credits is the snapshot figure. It is not cheap enough to treat as a sketch pad for every length on the ladder. Pick 2s, 3s, or 4s when that is the cut. Climb only when the action needs the climb.
Best Wan model and best text-to-video model exist so you can leave this row when the job leaves it. Stay when you need high-fidelity text-to-video, 1080p, a 2-second legal floor, optional audio in, 15 credits.
FAQ
What makes Wan 2.7 Text to Video different from other Wan rows?
This one is text-to-video, no still required, 2–15s, 1080p, 15 credits, optional custom audio input, prompt rewriting. Image-to-video Wan is a different contract.
Can I start at 2 seconds?
Yes. The catalog lists 2s as the first duration. Use it for hits and cutaways. Do not start at 15s out of habit.
Does it write a soundtrack if I bring nothing?
The audio flag is empty. The description offers optional custom audio input. Bring a track or plan the mix later. Do not assume a native score.
Should I write a short prompt and let rewriting handle it?
No. Rewriting is listed; it is not a brief. Specify the shot. Let the model tidy, not invent the city.