A one-line prompt becomes a hero video reel via the Wan 2.7 API in minutes — we ship campaign teasers without storyboarding.
The Wan 2.7 API generates a 720p or 1080p video clip from a natural-language prompt (Alibaba Wan 2.7). The Wan 2.7 API supports 2-to-15-second outputs at 30 FPS, seed lock, enable_prompt_expansion auto-enrichment, and negative prompts — all through one REST endpoint, metered per second at the official rate.
Input
The text prompt sent to the model
Output
SampleRun the model to replace this previewWan 2.7 API: Text to Video
Use the Wan 2.7 API to generate a 720p or 1080p video clip from a natural-language prompt.
The Wan 2.7 API generates a 720p or 1080p video clip from a natural-language prompt (Alibaba Wan 2.7). The Wan 2.7 API renders 2-to-15-second clips at 30 FPS with configurable seed lock, prompt-expansion auto-enrichment, and negative prompts — all through one REST endpoint.
The Wan 2.7 API is metered per second at the official rate: $0.10/s at 720p, $0.15/s at 1080p — $0.50 per 5-second 720p clip (default), up to $1.50 at 10s/1080p. One RouterBase key works alongside 200+ other models — no provider-specific SDK.
Six reasons teams ship on the Wan 2.7 API
From prompt-only generation to seed-locked reproducibility — what the Wan 2.7 API gets right.
Natural-language video
Describe the scene in plain English; the Wan 2.7 API renders a 2-to-15-second clip from the prompt — no storyboard, no reference image required.
720p / 1080p, 30 FPS
Switch resolution per Wan 2.7 API call. Default 720p for fast iteration; 1080p for delivery. Pass an aspect_ratio parameter to choose framing.
2 to 15-second clips
Set output duration per call. The Wan 2.7 API charges linearly per second — cost scales smoothly with the duration you pick.
Seed lock
Pass a seed value and the Wan 2.7 API reproduces the exact same clip for the same prompt — deterministic output for A/B and regression testing.
Prompt expansion
Toggle enable_prompt_expansion and the Wan 2.7 API auto-enriches a short brief into a fuller scene description before generation.
Metered per second
The Wan 2.7 API charges at the official per-second rate — $0.10 at 720p, $0.15 at 1080p. Pay only for the seconds you render.
Quickstart
Get started with the Wan 2.7 API in 3 steps
From sign-up to your first Wan 2.7 API call in under 5 minutes.
Create a RouterBase API key
Sign up and generate sk-rb-… — one key works across the Wan 2.7 API and 200+ other models on RouterBase.
POST the prompt
Send the prompt plus resolution and duration in JSON to the Wan 2.7 API endpoint. The call is async — RouterBase polls upstream until the render completes (typically 30 s–3 minutes), then returns a signed URL.
Inspect the result
Each Wan 2.7 API response includes generation metadata for auditing and a render URL ready to embed in your product.
Pay only for what you use
RouterBase passes through partner-tier pricing. Compared against the model's official published API rate.
| Resolution | RouterBase |
|---|---|
| 1280*720 / 720*1280 / 960*960 / 1088*832 / 832*1088 5s | $0.500 |
| 1920*1080 / 1080*1920 / 1440*1440 / 1632*1248 / 1248*1632 5s | $0.750 |
About $0.150 per second.One run at the default settings (1920*1080 / 1080*1920 / 1440*1440 / 1632*1248 / 1248*1632) costs $0.750.With $1 you can run this model approximately 1 times.
What teams build with the Wan 2.7 API
Real production loads running on the RouterBase model catalog across 200+ models.
We swapped three providers through RouterBase in one line of code when OpenAI had its outage. Our users never noticed.
Prompt-only generation on the Wan 2.7 API lets our marketing team draft ad concepts without touching a frame editor — they hand us prompts, we ship clips.
enable_prompt_expansion on the Wan 2.7 API takes our terse prompts and fills out the scene — fewer creative bottlenecks, more variation per brief.
Plugged the Wan 2.7 API into my existing OpenAI SDK with one base URL swap — less work than writing this testimonial.
RouterBase puts 200+ models behind one key. Our research team stopped asking for new provider accounts.
Seed lock on the Wan 2.7 API gives us deterministic outputs for regression testing — same prompt, same seed, same clip every time.
We migrated 42 services to RouterBase over a weekend. The only PR comment we got was 'wait, that's all?'.
Per-second metering on the Wan 2.7 API means we pay only for the seconds we render — at our scale that's cleaner budgeting than per-clip pricing.
Frequently Asked Questions
Common questions about the Wan 2.7 API.
The Wan 2.7 API is RouterBase's pass-through to Alibaba Wan 2.7 text-to-video. It generates a 720p or 1080p clip from a natural-language prompt — 2–15 second durations at 30 FPS, seed lock, prompt-expansion, and negative prompts.