Wan AI 2.5 Image-to-Video: Image to Video Online
Bottom line
Wan 2.5 Image-to-Video animates a photo into a 5 or 10 second clip at 720p or 1080p; 65 credits for 5 s at 720p, priced above the newer Wan models. Upload a starting frame, choose Wan AI 2.5 Image-to-Video in the Image to Video tool, and animate it on VdoBloom. The credit price appears before every generation.
Wan AI 2.5 Image-to-Video sample clips
Every clip below is a real render made on VdoBloom. Each caption names the model that produced it and gives its length and size.
Wan AI 2.5 Image-to-Video — model preview
A 10-second landscape clip generated on VdoBloom with Wan AI 2.5 Image-to-Video: the preview shown when choosing this model in the VdoBloom picker.
10 s · 1280×720
Preview clip, generated on VdoBloom with Wan AI 2.5 Image-to-Video
Make one like this →Wan 2.5 Image-to-Video is the older of Alibaba's Wan photo-to-video models on VdoBloom. You upload a photo, write a prompt, and pick 5 or 10 seconds at 720p or 1080p; there is no text-only mode and no last-frame option.
It costs 13 credits a second at 720p and 23 at 1080p: 65 credits for 5 seconds at 720p, 130 for 10; 115 and 230 at 1080p. That is about twice what Wan 2.7 Image-to-Video or Wan 3.0 charge for a comparable clip, so for new work those are usually the better first try. Wan 2.5 stays useful when a project was built on it and later clips should match earlier ones. The 10 free signup credits do not reach its 65-credit minimum; a one-time 300-credit pack is $19.50 and never expires.
The clip above is the model-picker preview, a 10-second 720p clip generated on VdoBloom with Wan 2.5. Below are the specs, worked costs, the comparison table with Wan 3.0, Wan 3.0 Prime, Wan 2.7 and Seedance 2 Fast, where it runs, prompting notes and limits.
Wan 2.5 Image-to-Video specs at a glance
From VdoBloom's Wan 2.5 endpoint, its KIE service wrapper, the dashboard model settings and the credit table.
| Spec | Wan 2.5 Image-to-Video on VdoBloom |
|---|---|
| Clip length | 5 or 10 seconds |
| Resolutions | 720p, 1080p (1080p is the default) |
| Aspect ratio | No ratio setting; only your photo is sent, so the framing comes from it |
| Inputs | One photo plus a prompt, both required |
| Prompt | The first 800 characters are sent; prompt expansion is on by default |
| Audio | Not listed as native audio in VdoBloom's model config |
| Price per second | 13 credits at 720p, 23 at 1080p |
| Cheapest render | 65 credits: 5 seconds at 720p |
| Content filter | Flexible (VdoBloom's most permissive tier); platform rules still apply |
| Where it runs | Image to Video tab only |
What Wan 2.5 Image-to-Video costs: worked examples
Four prices cover every setting. Dollar figures use the Lite plan rate of $15 for 300 credits ($0.05 a credit); a one-time 300-credit pack is $19.50 and never expires, and subscription credits reset each billing cycle. The right-hand column shows the Wan 2.7 Image-to-Video price for the same length and resolution.
| Setting | Wan 2.5 | Wan 2.7 Image-to-Video |
|---|---|---|
| 5 s, 720p | 65 cr ($3.25) | 32 cr |
| 10 s, 720p | 130 cr ($6.50) | 64 cr |
| 5 s, 1080p | 115 cr ($5.75) | 48 cr |
| 10 s, 1080p | 230 cr ($11.50) | 95 cr |
Read more: Credit packs and plans · Cheapest AI video models by credits
What 300 credits buys: Wan 2.5 vs Wan 2.7 Image-to-Video
Whole clips from one 300-credit pack ($19.50 one-time, or the $15 Lite month). The same photo-to-video job fits roughly twice as many times on Wan 2.7.
| Setting | Wan 2.5: credits / clips | Wan 2.7: credits / clips |
|---|---|---|
| 5 s at 720p | 65 / 4 | 32 / 9 |
| 10 s at 720p | 130 / 2 | 64 / 4 |
| 5 s at 1080p | 115 / 2 | 48 / 6 |
| 10 s at 1080p | 230 / 1 | 95 / 3 |
Settings the endpoint accepts but the dashboard does not show
VdoBloom's Wan 2.5 endpoint takes a few fields beyond what the Image to Video tab sends. The tab sends your photo, prompt, length and resolution; everything else stays at its default.
| Field | Default | What it does |
|---|---|---|
| Prompt expansion | On | Lets the provider expand the prompt before generating |
| Negative prompt | None | Things to avoid; up to 500 characters are sent |
| Seed | None (random) | A fixed number for repeatable results |
| Resolution | 1080p | 720p or 1080p; anything else is refused |
| Length | 5 s | 5 or 10 s; anything else is refused |
Wan 2.5 vs Wan 3.0 vs Wan 3.0 Prime vs Wan 2.7 vs Seedance 2 Fast
On every shared setting, Wan 2.5 is the highest-priced Wan photo animator here: 65 credits for 5 seconds at 720p, against 32 on Wan 2.7 Image-to-Video and Wan 3.0 and 51 on Wan 3.0 Prime. It also has the shortest length menu (5 or 10 seconds). Where it differs in kind is the prompt handling: a required prompt with provider-side prompt expansion, and API-level negative prompt and seed fields.
Prices are VdoBloom credits from the live credit table; Seedance 2 Fast figures are its text or photo rate (no reference video). "Fixed last frame" means the model is offered where the end frame is pinned: Keyframe Morph and Seamless Loop accept only models whose service takes both a first and a last frame.
| Model | Starts from | Clip length | Resolutions | Audio | 5 s at 720p | Cheapest render | Fixed last frame |
|---|---|---|---|---|---|---|---|
| Wan 3.0 | Text, or a photo as the first frame | 2-30 s, every second | 480p, 720p, 1080p | Native, on by default | 32 cr | 7 cr (2 s, 480p) | Yes, in Keyframe Morph and Seamless Loop |
| Wan 3.0 Prime | Text, or a photo as the first frame | 2-30 s, every second | 480p, 720p, 1080p | Native, on by default | 51 cr | 10 cr (2 s, 480p) | Not offered in the frame tools |
| Wan 2.7 Image-to-Video | A photo (prompt optional) | 5, 10 or 15 s | 720p, 1080p | Not listed as native audio | 32 cr | 32 cr (5 s, 720p) | Yes, in Keyframe Morph and Seamless Loop |
| Wan 2.5 Image-to-Video | A photo plus a prompt | 5 or 10 s | 720p, 1080p | Not listed as native audio | 65 cr | 65 cr (5 s, 720p) | No |
| Seedance 2 Fast | Text, a photo, or first + last frame | 4-15 s, every second | 480p, 720p | Native | 50 cr | 19 cr (4 s, 480p) | Yes |
Read more: Wan 2.7 Image-to-Video: about half the credits · Wan 3.0: up to 30 seconds with native audio · Wan 2.6 Flash: 20 credits for 5 s at 720p · Wan 2.2 Animate Replace: swap a character into existing video
When Wan 2.5 is still the right pick
The case for Wan 2.5 is continuity, not price. If earlier clips in a campaign or series were made with Wan 2.5, switching models halfway changes the look of the motion, and a few extra credits per clip can be cheaper than re-rendering the clips you already shipped. It is also a fair second opinion: run the same photo through Wan 2.5 and Wan 2.7 at 5 seconds and 720p (65 + 32 credits) and keep whichever motion suits the shot.
For new work with no such constraint, start with Wan 2.7 Image-to-Video or Wan 3.0; both cover 720p and 1080p for less, and Wan 2.7 and Wan 3.0 also offer lengths past 10 seconds.
Where to run Wan 2.5 Image-to-Video on VdoBloom
Wan 2.5 is on the Image to Video tab of Video Creation only: pick Wan AI, then 2.5 Image-to-Video, upload a photo, write a prompt, and choose 5 or 10 seconds at 720p or 1080p. It is not offered on the Text to Video tab, and it is not one of the first/last-frame models in Keyframe Morph or Seamless Loop.
Read more: Wan 2.5 and 2.6 on VdoBloom: Flash, text and image to video
Prompting Wan 2.5 Image-to-Video
Only the first 800 characters of the prompt reach the model, and the provider's prompt expansion is on, so short, specific prompts work best: say what moves and how the camera behaves, and let the photo carry the rest.
- Keep the prompt under 800 characters; anything past that is cut before the request is sent.
- Describe the motion and the camera, not the scene the photo already shows.
- Use a clean, well-lit photo with one clear subject; crop it to the shape you want, since there is no ratio setting.
- Use 5 seconds for one move and 10 seconds for a move with a clear end; there is nothing in between.
- Test at 720p (65 credits) before the 1080p final (115 or 230 credits).
Read more: Image-to-video best practices
Limits and honest caveats
What Wan 2.5 Image-to-Video does not do on VdoBloom, from its endpoint and service code.
- Photo required; no text-only mode.
- Only 5 or 10 seconds, at 720p or 1080p; no 480p and no 4K.
- No last-frame input, so it is not offered in Keyframe Morph or Seamless Loop.
- No aspect-ratio setting; the framing comes from your photo.
- Prompts are cut to 800 characters before sending.
- Native audio is not listed for this model in VdoBloom's config.
- The highest price per second among the Wan photo-to-video models compared here.
- Failed generations are refunded automatically.
What creators use Wan AI 2.5 Image-to-Video for
- Matching earlier Wan 2.5 clips in an ongoing project, so new shots look like the ones already delivered.
- Animating a single product photo or portrait into a 5 or 10 second clip at 1080p.
- Side-by-side tests against Wan 2.7 Image-to-Video on the same photo before committing to a batch.
How to use Wan AI 2.5 Image-to-Video on VdoBloom
- 1Open the Image to Video tool and select Wan AI 2.5 Image-to-Video from the model picker.
- 2Upload your starting image and describe the motion you want in the prompt.
- 3Choose your duration (5–10s), quality (720p/1080p), review the credit cost, and generate.
Frequently asked questions
How much does Wan 2.5 Image-to-Video cost?
13 credits a second at 720p and 23 at 1080p: 65 credits for 5 seconds at 720p, 130 for 10; 115 and 230 at 1080p.
Do the 10 free signup credits cover a Wan 2.5 clip?
No. The cheapest Wan 2.5 job is 65 credits. Wan 3.0 can be tried within the free credits (7 credits for 2 seconds at 480p).
Can Wan 2.5 make video from text alone?
No. It needs a photo and a prompt. For text-only video use Wan 3.0 or Wan 2.7 Text-to-Video.
What lengths and resolutions does Wan 2.5 support?
5 or 10 seconds, at 720p or 1080p. The dashboard defaults to 1080p.
Is there a prompt length limit?
Only the first 800 characters are sent to the model. Prompt expansion is on by default, so a short prompt is fine.
Wan 2.5 or Wan 2.7 Image-to-Video?
Wan 2.7 costs about half as much (32 vs 65 credits for 5 seconds at 720p), adds a 15-second option and works with first and last frames in Keyframe Morph and Seamless Loop. Pick Wan 2.5 mainly to match clips already made with it.
Does Wan 2.5 Image-to-Video generate sound?
Native audio is not listed for it in VdoBloom's configuration. Wan 3.0 and Wan 2.6 Flash list native audio.
Can I set the aspect ratio?
No. Only the photo is sent, so the framing comes from it. Crop the photo before uploading.
Can I use Wan 2.5 for morphs or loops?
No. Keyframe Morph and Seamless Loop need a model that takes a last frame; Wan 2.5's request has no last-frame field. Use Wan 2.7 Image-to-Video or Wan 3.0 there.
Can I set a negative prompt or a seed for Wan 2.5?
The endpoint accepts both (a negative prompt of up to 500 characters and a fixed seed), but the Image to Video tab does not show them; it sends the photo, prompt, length and resolution, with prompt expansion left on.
Wan AI 2.5 Image-to-Video guides, prices and comparisons
This page is the Wan AI 2.5 Image-to-Video hub. Each guide below goes deeper on one question and links back here.
Wan 2.5 and 2.6 guide
Flash, text and image to video on VdoBloom.
Best AI model for image to video
Photo-to-video models compared.
Image-to-video best practices
Photos and prompts that animate well.
Cheapest AI video model
Credit costs across the catalog.
Wan vs Seedance
Which model family fits which job.
Wan AI review (2026)
The Wan family on VdoBloom, reviewed.