Wan 2.7 Text to Video: Full Model Guide
Wan 2.7 text-to-video (wan2.7-t2v) converts written descriptions into finished video clips — and gives you the longest prompt budget and finest controls on txtovid. Select it from the model dropdown at the top right of the generator and direct scenes with up to 5,000 characters, a negative prompt and a reproducible seed.
Specifications at a glance
- Model ID: wan2.7-t2v
- Input: text prompts up to 5,000 characters, any language
- Optional: negative prompt (up to 500 characters) and seed
- Output: 720P ($0.10/s list) and 1080P ($0.15/s list) + flat $0.05/s platform fee
- Aspect ratios: 16:9, 9:16, 1:1
- Durations: 3–15 seconds — typical render time 1–5 minutes
The 5,000-character advantage
HappyHorse text-to-video caps prompts at 2,000 characters on txtovid — Wan 2.7 more than doubles that. The extra budget pays off in two ways: richer single scenes (layer subject, action, setting, lighting, camera and style without cutting corners) and multi-shot videos, where you script each cut explicitly: "Shot 1 [0–3 seconds]: drone reveal over a foggy coastline. Shot 2 [3–8 seconds]: close-up of waves hitting rocks in slow motion."
The negative prompt handles the other half of direction — what should NOT appear. Common picks: "blurry, low quality, text, watermark, extra limbs, distorted faces". And the optional seed makes iteration scientific: fix it, change one word, and see exactly what that word does.
Pricing and how to select it
The list price is $0.10/second at 720P and $0.15/second at 1080P, plus the flat $0.05/second platform fee — so a 5-second 720P clip costs $0.75, roughly 21% less than the same clip on HappyHorse ($0.95). The exact total is shown before you generate; public sharing saves 10% and plans 2–7% more.
To use it, open the text-to-video generator and select Wan 2.7 from the model dropdown at the top right — HappyHorse 1.1 is preselected. The prompt field expands to 5,000 characters and the negative prompt and seed inputs appear automatically.