Two roles, two separate photos
Upload a clear photo for each person. The first takes the left role and the second takes the right, with a swap control if you change your mind.
Bimg AI / Duet studio 01
Bimg AI / Cast your duet
Add two photos to create your 10-second Hotel Lobby take.
JPG, PNG, WebP or BMP · Up to 20 MB
JPG, PNG, WebP or BMP · Up to 20 MB
Photo 01 replaces the left performer. Photo 02 replaces the right.
This clip supplies the stage and choreography for your cast.
Give the orange-studio duet a new cast. Add one photo for each performer, set your video format, and create a 10-second Hotel Lobby take with Bimg AI.
Ideas for your next duo
Play these examples with sound to see how a new pairing changes the performance. Use them as casting inspiration for your own take.
Two grandmothers bring a new character to the orange-studio performance.
A familiar-looking performer shares the microphone with an unusual partner.
Coordinated gold outfits turn two characters into a visually connected duo.
01 / What makes the duet
The fixed Hotel Lobby clip sets the scene and performance. Your photos define the cast; the workspace handles the model settings and generation instructions.
Upload a clear photo for each person. The first takes the left role and the second takes the right, with a swap control if you change your mind.
The orange backdrop, hanging microphone, and shared dance gestures provide the visual reference for each new duet.
Choose an aspect ratio and 480P, 720P, or 1080P resolution. The clip length stays fixed at 10 seconds.
The displayed credit total follows Wan 3.0 Standard pricing. Preview the completed take, then download it or find it in your history.
02 / Build your cast


Casting ideas / AI concept artwork
The workspace needs two photos and two format settings. The performance template and generation instructions are already prepared.
Use sharp photos with visible faces and some clothing detail. Avoid group shots, covered faces, and extreme crops.
Check the photo labels before generating. Use Swap performers to move each photo to the other side.
Choose a ratio and resolution, check the credit cost, and generate. Review both faces, movement, and framing before downloading.
03 / Before you generate
Photos, video settings, credits, and what to check in your finished duet. You can also visit Hotel Lobby AI.
It creates a new 10-second duet using a fixed orange-studio performance and your two character photos. The look is inspired by the Hotel Lobby performance on A COLORS SHOW, with two performers sharing a hanging microphone.
Upload one separate photo for each role. Photo 01 replaces the performer who starts on the left, and Photo 02 replaces the performer who starts on the right. You can swap the photos before submission.
Choose one clearly visible person per image, with a sharp face and useful clothing detail. Each image must be JPG, PNG, WebP, or BMP and no larger than 20 MB.
This workspace uses a fixed performance template and prepared replacement instructions. You only choose the two photos, aspect ratio, and resolution. Open the Wan 3.0 workspace for a more flexible video brief.
The instructions ask the model to keep the original studio and performance while replacing both people. AI output can vary, so check both identities, motion, and the ending before sharing.
Each generation produces a 10-second video. Wan 3.0 Standard pricing is 80 credits at 480P, 160 credits at 720P, and 320 credits at 1080P. A paid Bimg AI account and enough credits are required.
Yes. Choose a supported aspect ratio, including 9:16 for vertical video. Original ratio follows the template. Reframing can affect how much of the stage is visible, so review the composition.
The task is saved to your account history. You can check it there during generation, return to the workspace, and download the completed video. A new generation is a separate task with its own displayed credit cost.
Your next Bimg AI take
Two photos are all your cast needs. Choose the format, check the cost, and create your own 10-second Hotel Lobby performance.
Start castingSwitch tools without breaking your creative flow.
Supports first/last frame video, multi-image reference generation
Create synchronized videos up to 30 seconds with native audio, multimodal references, and precise timeline control
Create Wan 3.0 and Wan 3.0 Prime videos from text, images, video, or audio references. Generate 2–30s clips with sound in 480P, 720P, or 1080P.
Multimodal AI video with text, image, video, and audio references
Generate MiniMax H3 videos from text, frames, or multimodal references in 768P or 2K with synchronized stereo audio on Bimg AI.
Create Kling 4.0 videos on Bimg AI with cinematic prompts, image references, shot planning, and free credits for new users.
Create AI videos with Kling 3.0 on Bimg AI using prompts, element references, multi-shot planning, and available audio settings.
Native 1080p video with synchronized audio and lip sync
Language
17 languages available
