WAN 2.1 Video
Generated on PixelDojo with WAN 2.1 Video. Produced by PixelDojo's generation pipeline.
WAN 2.1 Video costs 1.5 credits for every second of clip you request, so the cheapest possible run, 1 second, is 1.5 credits. Ours took about 3 minutes 44 seconds on August 26, 2026 to animate a product photo, and the file that came back was not the aspect ratio we asked for. This page walks through the per-second formula, what our run actually returned, and when the newer WAN 2.2 Standard is the better 1.5-credit-plus choice instead.
The 1.5 credit, 1 second clip
Every example below was produced on PixelDojo. Hover to see the prompt.
Slow cinematic camera orbit around the earbuds case, soft studio lighting shifts, the lid opens gently to reveal the earbuds, product commercial style
WAN 2.1 Video
What the run showed
The price is duration times 1.5, nothing else
Resolution (480p or 720p) does not change the charge. Only the duration you send, clamped 1 to 5 seconds, does. A 5 second clip is 7.5 credits at either resolution.
Our 1 second run: 1.5 credits, about 224 seconds
We sent the earbuds product photo with a camera-orbit prompt at 480p, duration 1. It charged 1.5 credits and took roughly 3 minutes 44 seconds end to end on Replicate's wan2.1-with-lora backbone.
Requested 16:9, got a square
We set aspect_ratio to 16:9, but the file we downloaded measured 640 by 640, matching our square 1024 by 1024 source image, not the requested ratio. In image-to-video mode the model appears to inherit the source image's shape rather than the aspect_ratio field.
It carries a LoRA slot text-to-video models rarely offer
lora_url takes a HuggingFace, CivitAI, or .safetensors link with an adjustable lora_strength, on top of the same text-and-image contract.
Why Choose Pixel Dojo for WAN 2.1 Video
Professional-quality results with cutting-edge AI technology
1.5 credits per second, flat
No resolution surcharge. 480p and 720p cost the same at any given duration.
1 to 5 second range
The shortest allowed clip, 1 second, is also the cheapest at 1.5 credits.
Custom LoRA support in the same request
Point lora_url at a hosted safetensors file and it applies at generation time, no separate fine-tuning step.
How It Works
How we ran it, so you can repeat it:
Send an image and a motion prompt
image_url plus prompt switches the tool into image-to-video mode. Leave image_url off and it runs text-to-video instead.
Set duration to control both length and price
1 to 5 seconds at 1.5 credits per second. We used 1 second, the floor, for 1.5 credits.
Check the file you get back, not just the setting you sent
Ours requested 16:9 and returned a 640 by 640 square. Confirm the real dimensions with the file itself before building a layout around a requested aspect ratio.
Loved by creators on PixelDojo
Real feedback from people using PixelDojo, pulled from our in-product surveys.
A "one-stop-shop" for creators! Thanks!!
Lots of different tools. It's easy to purchase more credits.
Amazing performance!
Excellent website for creating all types of media
Very eay to use, works well to train SDXL loras.
The amazing tools
Explore more AI tools on PixelDojo
AI Tools
Compare & Switch
- Best AI Image Generators
- Best AI Video Generators
- Midjourney Alternatives
- Civitai Alternatives
- Runway Alternatives
- Leonardo Alternatives
- Pika Alternatives
- Luma Alternatives
- Magnific Alternatives
- Veo Alternatives
- Flux Alternatives
- Freepik Alternatives
- Seedance Alternatives
- Seedream Alternatives
- Pixverse Alternatives
- GPT Image Alternatives
- Synthesia Alternatives
- Playground Alternatives
- NightCafe Alternatives
- Canva AI Alternatives
- ElevenLabs Alternatives
- ComfyUI Alternatives
- Fal Alternatives
- Replicate Alternatives
Common Questions
Everything you need to know about WAN 2.1 Video
What does WAN 2.1 Video cost?
1.5 credits for every second of the clip, 1 to 5 seconds, at either 480p or 720p. The cheapest run, 1 second, is 1.5 credits. Five seconds is 7.5 credits.
How fast is WAN 2.1 Video?
About 3 minutes 44 seconds in our August 26, 2026 run for a 1 second clip. Real turnaround moves with queue load on the underlying Replicate backbone.
Does it respect the aspect ratio I request?
Not reliably in image-to-video mode. We asked for 16:9 and got back a 640 by 640 square that matched our square source image instead. Text-to-video mode, which has no source image to inherit from, is more likely to follow the setting.
How do I call it from the API?
POST to /api/v1/models/wan-2.1-video/run with your API key and a prompt. Add image_url to switch to image-to-video, and duration to set both length and price.
How does it compare to WAN 2.2 Standard?
WAN 2.2 Standard is a flat 3 credits at 480p regardless of duration, so a 5 second clip is 3 credits there against 7.5 credits on WAN 2.1. WAN 2.1's edge is the per-second floor for very short clips and its LoRA slot, which WAN 2.2 Standard does not carry.