01/09/2026
REFERENCE IMAGE IS THE HIGHEST PRIORITY.
Use the uploaded reference image exactly as it is.
Keep the exact same person, face, hairstyle, beard, skin tone, facial features, body shape, clothing, food, plate, container, table, background, lighting, camera angle, composition, and every visible object exactly as shown in the uploaded reference image.
Do not create a new person. Do not replace, modify, regenerate, recolor, enlarge, reduce, or transform any object. The uploaded reference image defines every visual detail. Only animate the existing scene naturally.
This video starts exactly where the previous video ended. The first frame must be visually identical to the last frame of the previous video with seamless continuity.
Ultra realistic cinematic 4K vertical mukbang video, 10–12 seconds duration, extreme close-up framing, premium mukbang setup, dramatic studio lighting, sharp focus, ultra detailed food textures, glossy oil reflections, shallow depth of field.
The person remains seated behind the table facing the camera. The plate, food, and container remain exactly as they were at the end of the previous video.
Action Sequence:
The video begins with the person naturally continuing to chew the same bite taken in the previous video. The same partially eaten piece of food remains in the person's hand with identical bite marks and missing portion. Nothing changes between the last frame of the previous video and the first frame of this video.
After a few natural chews, the person gently lowers the same partially eaten piece and places it back onto the plate. The bite marks remain clearly visible and the missing portion never regenerates or changes.
The plate still contains every piece of food that was poured in the previous video. Every food item stays exactly where it was unless the person physically touches it. No food disappears, resets, duplicates, changes size, changes color, or transforms into another food.
The person naturally reaches toward the plate, selects another visible piece of food from the plate, slowly lifts it toward the camera, briefly presents it, and takes one large satisfying bite. The bite looks completely realistic with detailed food texture, natural tearing, and visible juicy interior. The missing portion remains permanently missing.
The person continues chewing slowly and naturally while occasionally looking toward the camera. Hand movements, finger movements, blinking, breathing, facial expressions, and body movements remain smooth and lifelike.
Near the end of the video, the person gently places the same partially eaten piece back onto the plate. It remains partially eaten with the exact same bite marks. Nothing repairs itself, refills itself, or returns to its original shape.
The final frame must look like a natural continuation, ready for the next video without any jump cut or visual change.
Audio (Very Important)
No voice.
No talking.
No narration.
No background music.
Only realistic ASMR sounds:
- natural hand movement
- finger touch sounds
- plate movement
- realistic food handling
- juicy bite
- chewing
- natural food texture sounds
- subtle table sounds
Visual Details:
Hyper realistic food textures, realistic physics, smooth cinematic motion, premium mukbang style, stable camera, no zoom, very close framing, no text, no subtitles, no watermark, no visual glitches.
Maintain perfect identity consistency, perfect object consistency, perfect food continuity, perfect bite continuity, and perfect scene continuity throughout the entire video. The first frame must exactly match the previous video's last frame, and the last frame must naturally continue into the next video.
