AI video models · compared
Wan 3.0 vs Wan 2.5
Wan 3.0 and Wan 2.5 are both AI video models that run on one varg API key. This page puts what the varg catalog states about each side by side: the price, how it is billed, the limits of the input, what it takes and the typical time. It ranks neither on quality.
- Wan 3.0
- From $0.10 per run from $0.05/s
- Wan 2.5
- From $0.25 per run from $0.05/s
Every model is billed at its provider's price with no markup, from one credit balance: 1 credit is $0.01.
Wan 3.0 vs Wan 2.5: side by side
Every row is read from the varg catalog and each model's input schema. A "from" price is the cheapest request the catalog states for that model, at its own cheapest settings. For the exact price of a request, send its body to POST https://api.varg.ai/v2/estimate.
| Fact | Wan 3.0 | Wan 2.5 |
|---|---|---|
| Made by | Alibaba | Alibaba |
| Takes and makes | Text to video, Image to video, Video to video | Text to video, Image to video |
| Price | From $0.10 per run | From $0.25 per run |
| Rates on the cheapest route | Resolution 480p: $0.05/s; Resolution 720p: $0.10/s; Resolution 1080p: $0.20/s; Other: $0.20/s; Minimum charge: $0.10 | Resolution 480p: $0.05/s; Resolution 720p: $0.10/s; Resolution 1080p: $0.15/s; Minimum charge: $0.25 |
| Billed | per second | per second |
| Clip length | 2 to 30 s | 5 or 10 s |
| Resolutions you can set | 480p, 720p, 1080p | 480p, 720p, 1080p |
| Audio switch | No: the input schema says the model ignores "audio" | No: the input schema says the model ignores "audio" |
| Typical time | ~3 min | ~2 min |
| Routes | 6 on fal | 2 on fal |
| Model id | wan_3 | wan_2_5 |
Which to pick
Each line is a difference between two rows of the table. None of it says which model's output looks better: that depends on the prompt.
Pick Wan 3.0 when
- you want the lower starting price: from $0.10 per run, against from $0.25 per run, each at the model's own cheapest settings.
- you need a clip longer than 10 seconds: it takes up to 30 in one request.
- you need video to video: the catalog lists a route for it on Wan 3.0, none on Wan 2.5.
Pick Wan 2.5 when
- your request sets resolution 1080p: its cheapest route charges $0.15/s there, against $0.20/s.
- you want the shorter typical wait: ~2 min, against ~3 min, by the catalog's estimates.
Call Wan 3.0 and Wan 2.5
One endpoint and one API key for both: POST https://api.varg.ai/v2/tools/video/call with Authorization: Bearer $VARG_API_KEY. It answers 202 with a job right away; poll GET /v2/jobs/{id} until its status is completed.
Wan 3.0
curl -X POST https://api.varg.ai/v2/tools/video/call \
-H "Authorization: Bearer $VARG_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "wan_3",
"prompt": "A red paper lantern drifts up through a rainy night market in Hong Kong, neon signs reflecting in wet streets, steam rising from food stalls. The camera follows the lantern as it rises above the crowd and reveals the glowing city skyline. Sounds of rain, sizzling woks and distant chatter.",
"aspect_ratio": "16:9",
"resolution": "720p",
"audio": true,
"duration": 5
}'vargai call_tool --tool video '{"model":"wan_3","prompt":"A red paper lantern drifts up through a rainy night market in Hong Kong, neon signs reflecting in wet streets, steam rising from food stalls. The camera follows the lantern as it rises above the crowd and reveals the glowing city skyline. Sounds of rain, sizzling woks and distant chatter.","aspect_ratio":"16:9","resolution":"720p","audio":true,"duration":5}'Wan 2.5
curl -X POST https://api.varg.ai/v2/tools/video/call \
-H "Authorization: Bearer $VARG_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "wan_2_5",
"prompt": "cherry blossom petals falling in slow motion, Japanese garden",
"files": [
{
"url": "https://s3.varg.ai/u/1580f459-9459-47db-9400-b64100181743/tmp_3545c2a5-01d.png"
}
],
"aspect_ratio": "9:16",
"duration": 9
}'vargai call_tool --tool video '{"model":"wan_2_5","prompt":"cherry blossom petals falling in slow motion, Japanese garden","files":[{"url":"https://s3.varg.ai/u/1580f459-9459-47db-9400-b64100181743/tmp_3545c2a5-01d.png"}],"aspect_ratio":"9:16","duration":9}'How billing works
Every model is billed at its provider's price with no markup, from one credit balance: 1 credit is $0.01. varg's fee is on the top-up, not on the run: 5.5% of the credits you buy ($0.80 at the least), or 8% for a team workspace. Pricing
Wan 3.0 vs Wan 2.5: questions
- Which is cheaper, Wan 3.0 or Wan 2.5?
- Wan 3.0 starts lower: from $0.10 per run, against from $0.25 per run for Wan 2.5. Those are each model's cheapest settings, and the price of a request depends on what it asks for: the table lists each model's rates. With resolution 1080p, Wan 2.5 is lower: Wan 3.0 $0.20/s, Wan 2.5 $0.15/s, each on its cheapest route. For the exact price of a request before you run it, send the same body to POST https://api.varg.ai/v2/estimate.
- Which makes longer clips, Wan 3.0 or Wan 2.5?
- Wan 3.0: up to 30 seconds in one request, against 10 for Wan 2.5, by their input schemas.
- Which resolutions can I set on Wan 3.0 and Wan 2.5?
- Wan 3.0: 480p, 720p, 1080p. Wan 2.5: 480p, 720p, 1080p. These are the tiers the "resolution" field of each model's input schema lists; a model can also have a default size that is not one of them.
- Can I use Wan 3.0 and Wan 2.5 with one API key?
- Yes. Both run on varg behind POST https://api.varg.ai/v2/tools/video/call with one API key and one credit balance: send "model": "wan_3" or "model": "wan_2_5". Every model is billed at its provider's price with no markup, from one credit balance: 1 credit is $0.01.