Wan 3.0 Prime
Alibaba Wan 3.0 Prime - up to 30s from text, an image (with optional end frame) or up to 10 references, audio included. On Influverse a 5s clip at 480p costs 13 credits, charged from the same wallet as every other model. There is no separate subscription for this model and no training step to use your own character with it.
By Influverse AI Team Last updated: 6 September 2026 Editorial policy
Generate with Wan 3.0 PrimeCapabilities
| Vendor | Alibaba |
|---|---|
| Duration | 2 to 30 seconds |
| Resolutions | 480p, 720p, 1080p |
| Audio | Native audio on every generation |
| Text to video | Yes |
| Image to video | Yes |
| First and last frame | Yes |
| Motion control | No |
| Extend an existing clip past its last frame | No |
| Edit an existing clip from a prompt | No |
| Reference video | Yes |
| Reference images | Up to 10 |
What it costs in credits
Derived from the same pricing source the app charges against, so these figures and the cost preview on the generation screen can never disagree.
| Duration | Resolution | Credits |
|---|---|---|
| 2s | 480p | 6 |
| 5s | 480p | 13 |
| 10s | 480p | 26 |
| 30s | 480p | 78 |
| 2s | 720p | 11 |
| 5s | 720p | 27 |
| 10s | 720p | 54 |
| 30s | 720p | 160 |
| 2s | 1080p | 22 |
| 5s | 1080p | 54 |
| 10s | 1080p | 107 |
| 30s | 1080p | 320 |
Credits come from your plan and from credit packs. See the pricing page for the plan ladder.
When to use Wan 3.0 Prime
Wan 3.0 Prime is the model to reach for when you want the widest set of inputs at the lowest running cost, and it is the one we now put first in the picker for that reason. It will work from a prompt on its own, from a single image, from a start frame and an end frame in the same generation, or from a stack of references that can mix images, video and audio in one call. That last combination is the interesting one, because it lets you hand the model a photo of your character, a clip of the movement you want and a piece of audio to match, then describe in the prompt how the three fit together. It is also the only model in the catalogue that will make a clip as short as two seconds, which is the right length for a cut you are going to drop into a longer edit rather than post on its own. At the other end it runs to thirty seconds, and of the models that go that far it is the cheapest per second at its lowest resolution tier, so a long take stops being a decision about budget. There are three resolution tiers and the price roughly doubles at each step, so the working pattern is to draft at the lowest one until the prompt is right and only then rerun at the tier you actually want to publish. Audio comes back with the clip rather than being layered on afterwards, and it costs nothing extra. You can also leave the length alone and let the model choose it from your prompt and your references, which helps when you do not yet know whether a beat wants three seconds or eight. Reach for something else when you need to copy movement from a clip frame for frame, because that is a different job and Kling Motion Control is built for it.
Strengths
- Two second clips, which nothing else in the catalogue will make
- Runs to thirty seconds at the lowest cost per second of any model that long
- References can mix images, video and audio in one generation
- Start frame and end frame control without a second pass
- Audio generated with the clip at no extra charge
Limits
- Three resolution tiers with a large price step between each, so the tier is a real decision
- Will not take a finished clip as a starting point, so longer cuts are assembled in the editor
- Reference video and audio are capped at fifteen seconds in total, and a reference clip needs at least sixteen frames per second
Best for
- Short cuts of two or three seconds for a longer edit
- Long single takes where the cost per second matters
- Scenes built from a character reference plus a movement clip
Other models to consider
Back to the full model catalogue
More from Influverse
Generate with Wan 3.0 Prime
Every new account starts with a free trial that unlocks every surface and every model. No card required.
Start creating free