Decision snapshot
| Decision factor | Wan 3.0 | Veo 3.1 |
|---|---|---|
| Text-to-video Elo | 1,240 | 1,090 |
| Per-minute cost at 720p | $12.00 | $24.00 |
| Elo lead | +150 | — |
| Document input | Yes | No |
| Audio in single pass | Yes | Yes |
Where each model wins
- Wan 3.0 wins on arena rank and price — pick Wan 3.0 for production video.
- Veo 3.1 Lite is cheaper per minute — but scores 151 points lower on the arena.
- Document input — only Wan 3.0.
- Native 30-second single-pass — both ship 30s.
Cost on the same workload
A 30-second 1080p clip costs $6.00 on Wan 3.0 versus $12.00 on Veo 3.1. Veo 3.1 Lite is cheaper at $4.80 per minute but scores 151 Elo lower — the cost savings come with a visible quality drop.
Ready to test the workflow?
Create account & add creditsWorkload recommendation
| Workload | Recommendation |
|---|---|
| Production 1080p narrative | Wan 3.0 — 150 Elo ahead at half the cost. |
| High-volume teaser production | Veo 3.1 Lite — cheaper per minute, accept the quality drop. |
| Document-to-video | Wan 3.0 — only it accepts documents. |
| 30-second single-pass with audio | Both — pick by Elo preference. |
Benchmark rules
- Same Artificial Analysis harness version.
- Compare same-resolution tiers.
- Vendor-reported and independently measured results are separate classes.
- Do not extrapolate a single benchmark to production reliability.
Frequently asked questions
Can I try Wan 3.0 before integrating it?
Use the OneInfer Wan 3.0 launcher to open a prepared prompt in the public playground. Availability is checked against the current model catalog.
How should I treat benchmark claims?
Check the provenance label and harness version. Vendor-reported and independently verified results are deliberately shown as different evidence classes.
Put Wan 3.0 to work
Fund a controlled evaluation, send a reference frame or document, and measure quality and cost on your own workload.