Wan 2.1 Mocha

Wan 2.1 Mocha

Playground

MoCha performs Video-To-Video character swaps using reference images, replacing a video’s character without per-frame pose or depth maps. Ready-to-use REST inference API, no coldstarts, affordable pricing.

Features

MoCha 🎭 — AI Video Character Replacement

MoCha is an end-to-end video character replacement system that seamlessly swaps the main character in a video with a new one provided via reference images. Unlike traditional methods, it requires no explicit per-frame structural guidance (such as pose or depth maps), while maintaining realistic motion, lighting, and facial expressions throughout the clip.

🌟 Key Features

• 🧠 Structure-Free Replacement No need for pose or depth maps — MoCha automatically aligns motion, expression, and body posture.

• 🎥 Motion Preservation Accurately transfers the source actor’s motion, emotion, and camera perspective to the target character.

• 🎨 Identity Consistency Maintains the new character’s facial identity, lighting, and style across frames without flickering.

• ⚙️ Easy Setup Works with a single image and a source video — no need for complex preprocessing or rigging.

• 💡 High Realism, Low Effort Perfect for film, advertising, digital avatars, and creative character transformation.

💰 Pricing

Billing Rules

• Minimum charge: 5 seconds - any video shorter than 5 seconds is billed as 5 seconds.
• Maximum billed duration: 120 seconds (2 minutes)

⚙️ How to Use

1. Upload image — A clear reference image of the new character (recommended formats: JPG / PNG, avoid WEBP).
2. Upload video — The motion source; MoCha extracts pose and expression dynamics from this clip.
3. Add prompt (optional) — Guide the output, e.g. “preserve outfit; natural expressions; no background changes.”
4. Select resolution — Choose between 480p or 720p.
5. Generate — Wait a moment while MoCha processes the replacement.
6. Review & Iterate — Fix a seed to reproduce results, or vary it for A/B comparisons.

🧩 Tips for Best Results

• Match Pose & Composition: Keep your reference image’s camera angle, body orientation, and framing close to the target video.
• Keep Aspect Ratios Consistent: Use the same aspect ratio between your input image and video.
• Limit Video Length: For best stability, keep clips under 60 seconds — longer clips may show slight quality degradation.
• Lighting Consistency: Match lighting direction and tone between image and video to minimize blending artifacts.

# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/wan-2.1/mocha" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "resolution": "480p",
    "seed": -1
}'

# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v3/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"