Core capability
The scoreboard
Production window: the 21 days to 9 September 2026 unless a 60-day window is named. Rates are rounded down. Wan 2.2 Spicy image-to-video: 16,059 jobs, 658 creators, 98% finished, 0 review stops, median 44 seconds, average quote about 190 credits. Wan 2.7 Spicy image-to-video (60 days): 374 jobs, 95 creators, 86% finished, 0 review stops, median 139 seconds, average quote about 2,000 credits. Wan 3.0 reference-to-video: 6,901 jobs, 349 creators, 94% finished, 155 review stops (2.2%), median 210 seconds, average quote about 1,460 credits. Wan 3.0 image-to-video: 182 jobs, 91% finished, 6% review stops.
What each generation can do
Wan 2.2 Spicy: one still in, a short silent 720p loop out; the fastest and cheapest explicit route on the platform. Wan 2.7 Spicy: one still in, 720p or 1080p, 2 to 15 seconds, silent, no filter. Wan 3.0: text, image (with optional end frame) or up to 10 image, 5 video and 5 audio references in; 3 to 30 seconds at 480P to 4K tiers with native audio; no negative prompt; one upstream review of the rendered clip. Wan 2.7 standard routes add text-to-video and video editing without the Spicy label.
Why the block rate rises with capability
The Spicy routes are hosted image-to-video endpoints with the prompt filter removed and no output scan; there is nothing in the loop to refuse. Wan 3.0 runs the full official model through a provider that reviews finished videos, which is how it can offer audio, references and 30 seconds at official quality. Two percent is low, but it is not zero, and it concentrates on fast, multi-position sequences with faces turned away. Slow, continuous, face-visible scenes pass at close to the Spicy rate.
Versions and current access
All three routes are available in Makeartly Studio now; the CTA opens the Wan family picker where each is selectable.
Evidence
Cost per finished clip
Wan 2.2 Spicy: under 200 credits for a loop, and effectively every run finishes. Wan 2.7 Spicy: quote by resolution and duration; a 15-second 1080p clip sits around the 2,000-credit average. Wan 3.0: per second at 100, 220, 440, 660 or 990 credits by tier, so a 5-second 720P clip is 1,100 and a 30-second 1080P scene is 13,200. Failed and blocked runs are refunded on all three, so only finished clips count against the balance.
Deployment filtering and adult-content boundaries
Makeartly does not add a separate mainstream adult-content refusal layer for verified adult creators; this deployment-level statement does not mean the upstream official model is uncensored or guarantee prompt acceptance. Adult workflows are for adults only. CSAM and all under-18 sexual content are prohibited. Consent and likeness authorization are required, and non-consensual intimate content, unauthorized real-person deepfakes, and sexual impersonation are prohibited.
The one-line rule
Short silent explicit loop: Wan 2.2 Spicy. Explicit single-character clip up to 15 seconds at 1080p: Wan 2.7 Spicy. Anything with audio, an end frame, more than one reference, or more than 15 seconds: Wan 3.0, prototyped at 480P. Creators who route this way finish the most clips per credit pack and almost never see a review stop, because each route is used for what it does not refuse.
Use cases
You post several short clips a day
Wan 2.2 Spicy: 44-second median render, under 200 credits, no review stops across 16,059 jobs.
You produce one long, narrated scene a week
Wan 3.0 reference-to-video with a beat-by-beat prompt; budget per second and accept a roughly 2% chance of a refunded review stop.
- Write down the brief: length, sound or silent, one still or several references, explicit or suggestive.
- Under 5 seconds, silent, one still: Wan 2.2 Spicy.
- Up to 15 seconds, 1080p, one still, explicit: Wan 2.7 Spicy.
- Audio, end frame, multiple references or over 15 seconds: Wan 3.0 at 480P first, then the final tier.
- If a Wan 3.0 result is review-stopped twice, move that brief down one route rather than rewording it.
Limitations and decision checks
Acceptance test the same still on all three
Run one 5-second clip from the same still on Wan 2.2 Spicy, Wan 2.7 Spicy and Wan 3.0 as an acceptance test, compare motion, identity and quote side by side, stop on the route that blocked or drifted, and choose per brief from what you saw rather than from this page’s averages.
Our numbers are platform-wide; yours will depend on your stills and prompts. Averages hide the tails: The 2.2% Wan 3.0 stop rate is an average over 6,901 jobs of mixed content; explicit fast-motion briefs sit above it and slow scenes below it.
