Wan 2.7 R2V Tutorial
Practical Wan 2.7 reference-to-video workflow on Flixly. 8-second clips, 12-credit generations, and exact settings for consistent character motion.
TL;DR
Wan 2.7 R2V on Flixly turns one 1024x576 reference image into an 8-second 24 fps video for 12 credits. Set motion strength between 0.40 and 0.85. Chain results into Video to Video for extra polish. Follow the eight-step workflow above to keep character consistency across frames.
The real question behind Wan 2.7 R2V
Users search for Wan 2.7 R2V tutorials because they want consistent character motion from a single reference photo. The answer starts at the Reference to Video page. Upload one image, pick Wan 2.7, set duration to 8 seconds, and generate. That single path replaces hours of manual keyframing.
Direct answer first
Wan 2.7 accepts a 1024x576 reference image and outputs an 8-second 24 fps video clip. Motion strength sits at 0.65 by default. Change it to 0.45 for subtle head turns or 0.80 for full-body walks. Credit cost stays fixed at 12 credits per generation.
Layers most guides skip
Most tutorials stop after the first render. Wan 2.7 lets you chain outputs into Video to Video for style transfer. Run the first clip through Kling 3.0 at 0.3 strength to add film grain without breaking character consistency. Seedance 2.0 reference images work best at 512x512 square crops before upscaling.
Input size rules
- Reference image: 1024x576 or 1280x720
- Output resolution: locked to 1280x720
- Frame count: 192 frames at 24 fps
Tradeoffs nobody mentions
Wan 2.7 produces smoother motion than Image to Video but drops fine facial detail on profiles. Front-facing references retain 90 percent eye accuracy. Side profiles fall to 65 percent. If your project needs talking heads, route the clip to Lip Sync Video afterward.
Settings table
| Parameter | Default | Low motion | High motion |
|---|---|---|---|
| Motion strength | 0.65 | 0.40 | 0.85 |
| Guidance scale | 7.5 | 5.0 | 9.0 |
| Steps | 25 | 20 | 30 |
| Seed | random | fixed | random |
Step-by-step instructions
- Open the Reference to Video tool and sign in at /auth/register.
- Upload a 1024x576 PNG reference image that shows the full character.
- Select Wan 2.7 from the model dropdown.
- Set duration to 8 seconds and motion strength to 0.65.
- Enter a prompt such as "slow walk toward camera" and generate.
- Review the 192-frame clip at 24 fps.
- If needed, send the output to Video to Video and apply 0.3 strength with Seedance 2.0.
- Download the final MP4 or queue it for Auto Captions.
Decision rule
Choose Wan 2.7 when your reference photo must drive the entire motion sequence. Switch to First to Last Frame when you need precise start and end poses instead.
FAQ
What resolution works best with Wan 2.7? 1024x576 reference images produce the most stable 8-second clips. Wider ratios introduce edge warping after frame 120.
How many credits does one Wan 2.7 generation use? Each render costs 12 credits regardless of motion strength or guidance settings.
Can I use Wan 2.7 output in other tools? Yes. Export the MP4 and import it directly into Lip Sync Video or AI Video Effects for further edits.
Does Wan 2.7 support 16:9 aspect ratio? No. Output stays fixed at 1280x720. Crop or pad your reference image beforehand to match.
Frequently Asked Questions
What resolution works best with Wan 2.7?▾
1024x576 reference images produce the most stable 8-second clips. Wider ratios introduce edge warping after frame 120.
How many credits does one Wan 2.7 generation use?▾
Each render costs 12 credits regardless of motion strength or guidance settings.
Can I use Wan 2.7 output in other tools?▾
Yes. Export the MP4 and import it directly into Lip Sync Video or AI Video Effects for further edits.
Does Wan 2.7 support 16:9 aspect ratio?▾
No. Output stays fixed at 1280x720. Crop or pad your reference image beforehand to match.
