How to Use Wan in 2026: Access, Prompts, and Cost

To use Wan, open a text-to-video tool, describe the shot you want, choose Wan as the model, set your duration, and generate — the model returns a short, motion-rich 720p clip you can download or refine. You do not need a separate Alibaba account if you reach Wan through a multi-model platform: on Wandura's text-to-video tool it sits alongside other video models on one credit balance, so you pay per generation instead of a monthly per-model fee.

This guide covers the fastest way to access Wan, how to prompt it well, and exactly what it costs.

What Wan is good at

Wan is Alibaba's Wan (Wan2.2) video model, built to turn text and product images into believable motion. It shines on product showcases, social ad creative, and lifestyle scenes — animating a still product shot into a clean 720p clip without a photoshoot or an edit suite. It's also one of the most affordable video models on the platform, which makes it a natural pick for drafting and high-volume output where cost per clip matters. Because it lives next to other engines in one AI video generator, you can start on Wan and A/B the same prompt on Sora 2 or other models without switching apps — see what Wan does inside Wandura on the Wan model page.

How to use Wan, step by step

1. Open a text-to-video tool. On Wandura, go to text-to-video — no separate Alibaba account required.
2. Write a clear prompt. Name the subject, the action, the camera move, and the style in one line (see prompting tips below).
3. Select Wan as the model. Switch the model picker to Wan; the same prompt box works across models so you can compare later.
4. Set your duration. Choose anywhere from 4 to 15 seconds; keep the first clip short (4–5s) to iterate cheaply. Wan outputs 720p.
5. Generate and review. The clip is produced asynchronously; when it's ready, download it or tweak the prompt and re-run.
6. A/B against another model (optional). Re-run the same prompt on Sora 2 or another engine from the same balance and keep whichever nails the shot.

Prompting tips for Wan

Wan rewards specific, visual prompts. A useful pattern is subject + action + camera + lighting/style: "a matte-black water bottle on a wet stone surface, slow rotation, soft studio key light, shallow depth of field" beats "a water bottle". Describe motion explicitly ("slow dolly in", "gentle rotation", "handheld pan") and keep one main action per clip — stacking several actions into one prompt tends to muddy the result. For product and ad shots, name the setting and how you want the item to move so the motion feels intentional rather than random. Generate two or three variations before committing.

How much does Wan cost?

On Wandura, Wan is credit-based (1 credit = $0.01) and billed per generation — the exact cost is shown on the Generate button before you confirm, so nothing is hidden. Wan produces clips from 4 to 15 seconds at 720p, priced per second. Here's what a clip actually costs:

Prices as of July 2026 — the exact cost is shown on the Generate button; see pricing for current.

Wan is one of the most cost-effective video models on Wandura. For context, the same 5-second clip runs about 315–525 credits on Sora 2 depending on resolution — so Wan is a low-cost way to draft product motion and produce social clips at volume, all on the same balance, so you can A/B by model per shot.

Wan vs other video models

Wan is one of several models on the platform, and its low cost per clip makes it a strong fit for high-volume product and ad creative where you're generating many variations. When a shot needs frontier motion quality or synced native audio, it's worth A/B-ing against a premium engine. If you're deciding between engines, our Sora 2 vs Veo 3.1 vs Kling comparison breaks down motion, realism, audio, and cost — and the Wan model page shows exactly what Wan can do inside Wandura. All of them run on the same credit balance, so you can switch by shot instead of paying a separate subscription for each.