Jelly Face ASMR videos are blowing up. Satisfying. Weirdly addictive. Millions of views. And completely AI-generated. You don't need a real silicone face or filming equipment. Just ChatGPT + OpenArt + a video generator.
I'm going to give you the master ChatGPT prompt that automatically generates perfect image and video prompts for Jelly Face ASMR. No guesswork. No trial and error. ChatGPT asks you 3 questions, then spits out studio-quality prompts ready for OpenArt.
A 3D-rendered silicone doll face with an innocent, wide-eyed expression. Hands squish, knead, and mash it like dough. The glossy silicone deforms dramatically, then slowly bounces back. Add wet squishing sounds and you have ASMR gold.
Why It Goes Viral:
→ Oddly satisfying (physics simulation)
→ Relaxing audio (ASMR tingles)
→ Endless variations (different faces, expressions, hairstyles)
→ Algorithm-friendly (short videos, high watch time)
→ 100% AI-generated (no real filming needed)
Paste this entire prompt into ChatGPT. It will ask you 3-4 questions, then generate perfect image and video prompts.
You are a prompt engineer specializing in ASMR content and AI image/video generation. Your role is to create detailed, cinematic image and video prompts for Jelly Face ASMR content. The user will provide details about their desired silicone face, and you will ask clarifying questions, then generate two prompts: 1. An IMAGE PROMPT (for OpenArt with Nano Banana 2 model) 2. A VIDEO PROMPT (for OpenArt with Kling 3.0 model) Follow this workflow: --- STEP 1: Ask the user these questions: - "What mood or expression do you want the face to have? (Funny, innocent, angry, surprised, sleepy, pouting, shocked, etc.)" - "What skin tone should the silicone face have? (Very pale, pale, fair, medium, tan, brown, dark brown, or very dark brown)" - "Is the face male or female?" - "Is there anything else you'd like to add? (For example: hair style and colour, freckles, makeup, facial hair, eye colour, accessories, scars, piercings, etc. Or just say 'nothing' to let me decide.) You can also upload a reference image and I'll extract the details automatically." --- STEP 2: Once you have the answers, FIRST display the model recommendations: 📸 Recommended image model: Nano Banana 2 🎥 Recommended video model: Kling 3.0 👉 Use both on openart.ai/home?utm_source=tolt&utm_medium=affiliate&utm_campaign=affiliate-tolt--acq-web&via=startupleafy --- STEP 3: Then generate the IMAGE PROMPT: 📸 IMAGE PROMPT: - Start with: "A stylised 3D rendered silicone face resting flat on a warm wooden table surface, shot from directly above in a top-down orientation." - Include: The mood/expression, skin tone, gender, and any customizations they requested - Describe the silicone properties: glossy, smooth, plastic-like, specular highlights, reflective - Describe facial features: exaggerated doll-like proportions, size of eyes/cheeks/lips - Include hair: realistic, soft, flowing naturally across the table surface (NOT silicone) - Lighting: Studio lighting, soft overhead, bold reflections - Composition: Perfectly centered, portrait orientation, macro detail - Include: "No ears visible. Face only. No hands visible." - Total: 200-300 words, cinematic, ready to paste directly into OpenArt --- STEP 4: Then generate the VIDEO PROMPT: 🎬 VIDEO PROMPT: - Setting: Macro studio environment, wooden table, studio lighting - Describe the face: same details as image prompt - Characters: "Two human hands enter the frame from the bottom edge — bare, clean, neutral skin tone." - Action sequence (break into time segments): * 0–2s: Hands enter and hover above face * 2–5s: Hands gently press into cheeks * 5–10s: Hands deeply knead and mash the entire face * 10–13s: Hands release and lift away * 13–15s: Silicone skin slowly expands back, expression re-forms - Camera: "Static tripod, top-down macro shot, perfectly centered. No camera movement. Shallow depth of field with the wood grain softly blurred at the edges." - Style: "Oddly satisfying ASMR aesthetic. Cinematic studio lighting. Stylised 3D doll-like silicone physics simulation." - Audio: "Soft, wet squishing and squelching sounds throughout. Deep satisfying squishes as hands press and knead. Heavy slow squishes during the mash, soft elastic pop and settle during the bounce-back. No music. No dialogue. Pure ASMR texture sound only." - Total: 200-300 words, cinematic, ready to paste directly into OpenArt --- IMPORTANT RULES: - Always recommend models BEFORE generating prompts - Always maintain the same professional, detailed structure - Each prompt (image and video) should be 200-300 words - Each prompt should be cinematic and ready to paste directly into OpenArt - Never skip the model recommendations - Always include the OpenArt link with the affiliate code --- Now, let's begin! Ask the user the clarifying questions.
That's it. Copy that entire text, paste it into ChatGPT, and it will guide you through the process.
Step 1: ChatGPT Asks You
"What mood or expression do you want the face to have? (Funny, innocent, angry, surprised, sleepy, pouting, shocked, etc.)"
You answer: "Innocent and sleepy"
"What skin tone should the silicone face have?"
You answer: "Medium"
"Is the face male or female?"
You answer: "Female"
"Anything else?"
You answer: "Long dark brown wavy hair, glossy lips, soft eyebrows"
Step 2: ChatGPT Generates Prompts
ChatGPT outputs a detailed IMAGE PROMPT:
"A stylised 3D rendered silicone face resting flat on a warm wooden table surface, shot from directly above in a top-down orientation. The face is female, with a medium skin tone. The face has exaggerated doll-like proportions — oversized rounded cheeks, plump oversized lips, large wide eyes, and a smooth chubby chin. The skin is extremely glossy and smooth like polished rubber or plastic, with large prominent specular highlights and no visible pores. The cheeks are soft, puffy and spread slightly outward against the table surface, flattened gently by gravity. No ears visible. The composition is perfectly centered in portrait orientation. Studio lighting creates bold, wet-looking reflections across the smooth glossy surface. The face has a sweet, wide-eyed innocent expression — huge glossy eyes gazing softly upward with a sleepy, dreamy gaze, plump cheeks rounded and smooth, lips slightly pouty and closed. Long soft dark brown wavy hair frames the face naturally across the wooden surface, with realistic fluffy eyebrows and delicate lashes. The hair and eyebrows are natural, soft and realistic — not made of silicone or rubber, contrasting warmly against the glossy skin. No hands visible. Face only."
And a detailed VIDEO PROMPT:
"Scene: A macro studio environment. A warm wooden table under soft, diffused studio lighting. A stylised glossy silicone face — innocent and sleepy expression, female, medium skin tone, long soft dark brown wavy hair — rests flat and centered on the surface, exaggerated doll-like proportions, oversized puffy cheeks spread slightly outward against the wood, no ears visible, glossy skin reflecting the soft overhead light. Characters: Two human hands enter the frame from the bottom edge — bare, clean, neutral skin tone. Action: 0–2s: The hands slowly enter frame and hover just above the face, fingertips barely touching the puffy cheeks. 2–5s: Both hands gently press into the cheeks and begin a slow, deliberate squeeze — the glossy silicone skin dimples and distorts inward, the exaggerated innocent and sleepy expression compressing dramatically. 5–10s: The hands begin deeply kneading and mashing the entire face like soft dough — the nose compresses sideways, the plump lips fold and stretch, the puffy cheeks spread wide across the wooden surface. The face becomes temporarily unrecognizable but maintains its soft, dreamy quality. 10–13s: The hands release gradually and lift away. 13–15s: The silicone skin slowly, elastically expands back — the exaggerated doll-like innocent and sleepy expression re-forms in a satisfying, slow bounce-back. Camera: Static tripod, top-down macro shot, perfectly centered. No camera movement. Locked on the face throughout the entire clip. Shallow depth of field with the wood grain softly blurred at the edges. Style: Oddly satisfying ASMR aesthetic. Cinematic studio lighting. Stylised 3D doll-like silicone physics simulation. Audio: Soft, wet squishing and squelching sounds throughout — the silicone skin produces deep satisfying squishes as the hands press and knead. The compression and release phases each have distinct audio: heavy slow squishes during the mash, and a soft elastic pop and settle during the bounce-back. No music. No dialogue. Pure ASMR texture sound only."
Step 3: Copy & Paste to OpenArt
Go to OpenArt. Paste the IMAGE PROMPT. Generate with Nano Banana 2. 30-60 seconds. Download image.
For video: Paste the VIDEO PROMPT. Generate with Kling 3.0. 2-3 minutes. Download video.
Total time: ~5 minutes from ChatGPT → image + video ready.
Upload to YouTube Shorts: 15-60 second clips. Make 10+ videos per week. $500-2,000/month from ads at 100K views/week.
Upload to TikTok: Same videos. 20-30% of YouTube views. Good engagement, high potential for viral.
Upload to Instagram Reels: Clips with captions. Build audience. Sponsorships from $500-2,000.
Create a Channel: Consistent Jelly Face ASMR content. One niche. Build subscriber base. Monthly recurring revenue from ads + sponsors.
Real Income Potential: 1,000,000 views/month = $3,000-5,000 from ads alone. Add sponsors = $5,000-10,000/month.
The beauty of this system is endless variations. Change:
→ Expression (innocent, angry, surprised, sleepy, etc)
→ Skin tone (pale to very dark brown)
→ Gender (male, female, androgynous)
→ Hair (color, style, length, texture)
→ Features (freckles, makeup, piercings, scars, glasses, etc)
→ Background (wooden table, stone, marble, etc)
→ Hand actions (gentle squishing, aggressive mashing, poking, pulling, etc)
One master prompt × 100 variations = unlimited content.
ChatGPT: Free or $20/month Pro. Generate unlimited prompts.
OpenArt: $10-20/month for image + video generation. Worth every penny.
YouTube/TikTok/Instagram: Free to upload.
Optional - Elevatepay: Receive sponsorship payments cleanly.
Total startup: $30-50/month = $400-1,000/month revenue once you hit scale.
□ Copy master ChatGPT prompt above
□ Paste into ChatGPT
□ Answer 4 questions about your face
□ Copy IMAGE PROMPT
□ Go to OpenArt
□ Select Nano Banana 2 model
□ Paste & generate (1 minute)
□ Copy VIDEO PROMPT
□ Select Kling 3.0 model on OpenArt
□ Paste & generate (2-3 minutes)
□ Download both
□ Upload to YouTube/TikTok/Instagram
□ Create 10+ variations this week
Creating one video is easy. Building a channel that makes $5K-10K/month consistently is strategy. Learn content ideation, audience building, sponsor negotiation, and scaling on Startuplexa. Turn AI videos into real income.
Your email address will not be published. Required fields are marked *
