AI Crying Video: Create the Tears Effect From a Single Photo
Your photo transforms into a cinematic nighttime goodbye: facing someone special, tears glistening as you try to smile, the expression crumbling, then turning to leave.
The AI crying effect creates a 10-second cinematic moment from a single image. You're captured at night opposite a soft-focus figure, attempting a smile as tears well in reddened eyes, the expression faltering, tears streaming, before you pivot and disappear into darkness—all set to melancholy music. All you need is one good face photo.
Also called Cinematic Emotional Farewell · Crying trend · AI nangis · AI menangis · Filter wajah sedih
Step by step
How to make the AI Crying video on Smew AI
Everything happens in your browser. No editing software, no prompt writing, no watermark added.
Add your photo
Click "Create your own" and upload a clear face photo. Capture with your camera or select from your gallery.
Press Generate
The button displays the credit cost. You get a 10-second video directly from your photo, with its own audio. You don't write any prompt.
Download or share
Your video arrives within minutes as a 9:16 MP4. Save it, share directly from your device, or access it later in your Library.
Which photo works best?
Upload a clear, forward-facing image of your face with both eyes visible and open. The tears effect depends on eye detail.
A head-and-shoulders selfie is ideal. The video builds the nighttime setting around you.
Begin with a gentle smile or neutral expression. The video starts with a delicate smile.
Balanced lighting across your face. Heavy shadows or sunglasses obscure eyes and tears.
Background doesn't matter: the video replaces it with the night park.
Only use photos of yourself or adults who gave permission. Never strangers, celebrities or children.
Where did the AI Crying trend start?
The trend centers on a CapCut template called Cinematic Emotional Farewell. A tutorial from 25 September showed it had already hit 17.6 million uses. Every viral clip features the same composition: nighttime over-the-shoulder framing, a blurred figure in white ahead, warm bokeh behind, and a face transitioning from fragile smile to silent tears. Most versions then show the person walking off.
Indonesia and the Philippines saw the fastest adoption. An Indonesian tutorial on 25 September, "Cara buat filter wajah sedih, AI nangis di CapCut", reached 6.8 million views and 174,000 likes. Tagalog guides demonstrating the same in Dreamina AI appeared in October, hitting 688,000 and 678,000 views.
We analyzed the top 10 "AI crying" TikTok videos frame by frame using Gemini on 4 October and reconstructed the framing, performance and editing. The prompts here derive from that breakdown.
A "cry face" AI clip appears on TikTok, accumulating roughly 960,000 views; crying-effect content rises from this point. TikTok
Initial CapCut farewell videos with the nighttime over-the-shoulder framing emerge; one featuring a hijab hits 295,000 views. TikTok
An Indonesian guide for the template is published and surpasses 6.8 million views; the template displays 17.6 million uses in the video. TikTok
Tagalog guides demonstrate the same technique in Dreamina AI and achieve 688,000 views. TikTok
Every one of the 10 top "AI crying" TikTok videos features the identical over-the-shoulder night composition in our frame-by-frame analysis.
Copy and paste
The AI Crying prompts
These aren't required: the one-tap workflow handles them automatically. A photo prompt places your face into the scene, then a video prompt brings it to life. Want to modify them? Copy and adjust them on the Create page.
Farewell scene
1 photo: a clear face photoSeedance 2.5 Video Edit10 s · 720p · 9:16128 credits
Video prompt
Edit [Video 1]. Replace only the explicitly mapped source subjects in place:
[Image 1] replaces the crying foreground woman, including her departing back view; photo slot: A clear face photo, in every shot where that subject appears.
[Image 1]: Copy the uploaded person's face, facial structure, eyes, skin tone, age, hair and facial hair or absence of facial hair. Remove source facial hair or hairstyles that the upload does not have. Keep identity consistent across every shot. Keep the source outfit unless the selected variant explicitly changes it or requires a body/species change.
Selected variant: Farewell scene / best. Apply only the requested cast, wardrobe, setting or effect changes for this variant; preserve the source action and shot timing instead of inventing a new scene sequence.
Use the complete preview as the source for the AI Crying / Farewell scene variant. Preserve its actual action, effect progression, camera moves, shot sequence, physical interactions and mouth/jaw timing. Photo-role labels determine the intended identity assignment; never assign uploaded faces to unrelated background people or props.
Track each source subject through front, profile, back and occluded views. Keep all non-target subjects unchanged. Never duplicate an uploaded identity, merge subjects, swap their roles or transfer an identity to a prop or background subject.
Preserve the complete source video from its first frame to its last: source duration, frame order, every shot boundary, playback speed, camera movement, framing, lighting and original performance; keep background/props unless the selected variant explicitly changes them. Preserve each subject's original head, gaze, body, hand/paw gestures, expression, mouth and jaw opening/closing and their exact source timing. Do not create a new dance, scene, shot list, dialogue or lyrics. Do not add, remove, reorder, shorten, extend or retime shots or actions.
Produce silent video. The matching unmodified original soundtrack is restored separately after this edit. Do not generate speech, singing or music. Do not add new captions, subtitles, logos or a watermark.
1 photo: your selfieSeedance 2.5 Video Edit10 s · 720p · 9:16103 credits
Video prompt
Edit [Video 1]. Replace only the explicitly mapped source subjects in place:
[Image 1] replaces the crying foreground woman, including her departing back view; photo slot: Your selfie, in every shot where that subject appears.
[Image 1]: Copy the uploaded person's face, facial structure, eyes, skin tone, age, hair and facial hair or absence of facial hair. Remove source facial hair or hairstyles that the upload does not have. Keep identity consistent across every shot. Keep the source outfit unless the selected variant explicitly changes it or requires a body/species change.
Selected variant: Rainy night scene / rain. Apply only the requested cast, wardrobe, setting or effect changes for this variant; preserve the source action and shot timing instead of inventing a new scene sequence.
Use the complete preview as the source for the AI Crying / Rainy night scene variant. Preserve its actual action, effect progression, camera moves, shot sequence, physical interactions and mouth/jaw timing. Photo-role labels determine the intended identity assignment; never assign uploaded faces to unrelated background people or props.
Track each source subject through front, profile, back and occluded views. Keep all non-target subjects unchanged. Never duplicate an uploaded identity, merge subjects, swap their roles or transfer an identity to a prop or background subject.
Preserve the complete source video from its first frame to its last: source duration, frame order, every shot boundary, playback speed, camera movement, framing, lighting and original performance; keep background/props unless the selected variant explicitly changes them. Preserve each subject's original head, gaze, body, hand/paw gestures, expression, mouth and jaw opening/closing and their exact source timing. Do not create a new dance, scene, shot list, dialogue or lyrics. Do not add, remove, reorder, shorten, extend or retime shots or actions.
Produce silent video. The matching unmodified original soundtrack is restored separately after this edit. Do not generate speech, singing or music. Do not add new captions, subtitles, logos or a watermark.
You pay in credits. The Generate button shows this price before you spend anything. These are the options on this page.
Option
Model
What you get
Credits
Farewell scene: Best
Seedance 2.5 Video Edit
10 s · 720p · 9:16
128
Farewell scene: Standard
Seedance 2.5 Video Edit
10 s · 1080p · 9:16
56
Rainy night scene
Seedance 2.5 Video Edit
10 s · 720p · 9:16
103
One-time packs and plans are on the pricing page. A failed video gives its credits back.
What does the AI crying video look like, shot by shot?
We examined the top 10 AI crying TikTok clips frame by frame. All follow this structure:
Time
What you see
0 to 2 s
Nighttime over-the-shoulder close-up. A blurred figure in white shirt ahead. You gaze at them with reddened, glistening eyes.
2 to 4 s
You attempt a smile despite the tears. The smile wavers and collapses.
4 to 6 s
You blink, tears stream down each cheek, your chin trembles, your gaze drops.
6 to 10 s
A cut: you pivot and walk into the night, warm bokeh glowing behind.
A melancholy ballad plays throughout. The "Farewell scene" prompt here requests all four moments in one 10-second generation.
Why do people post AI crying videos?
Mainly for dramatic effect, often humorously. The clip delivers ten seconds of film-quality emotion on a recognizable face, making it perfect as a reaction: "me when the group chat leaves me on read", "me after one sad song", "when my favourite cafe closes". The contrast between trivial captions and K-drama intensity creates the humor. A Turkish caption in the trend captures it: "I wish I cried this beautifully in real life".
Some use it sincerely. Creators match the scene with breakup lines, moving stories or homesickness, and viewers respond in comments. The format requires only your face, explaining its rapid spread across languages.
Farewell scene or rainy night scene: which one should you make?
Farewell scene
Rainy night scene
Steps
Two: a photo prompt creates the night setting, then a video prompt animates it
Two, identical process
Length
10 seconds
8 seconds
Shot
Over-the-shoulder of a blurred figure, then cutting to you walking off
You alone on a rain-soaked neon street
Mood
The recognized trend format: a dramatic goodbye
Solitary, cinematic, no other person visible
Best for
Participating in the trend, reaction content
Story updates, photo series, longer text
Choose the farewell scene for the classic trend version. Select the rainy night scene when you want a solo moment without another person.
What goes wrong with the AI crying face, and how do you fix it?
Problem
Fix
The face changes into someone else
Upload a clearer, larger facial photo, then run it again.
The crying looks cartoonish or loud
Run it again. A second take is often calmer.
The person in front is too sharp
Run it again.
Makeup, a hijab or glasses shift
Use a clear photo where those are easy to see, then run it again. The template keeps clothes and glasses from the photo.
The video model refuses the photo
Some models reject certain real faces. Credits are refunded. Try another photo, or run it again.
Is it okay to make someone cry with AI?
Keep it lighthearted and respectful. The AI crying effect is dramatic performance, and the best posts are clearly playful. Create it with your own face or with an adult who consented and knows where it will appear.
Don't use it to fabricate someone's distress: no videos pretending a real person had a breakdown, no "evidence" in disputes, no content designed to manipulate or frighten someone. That causes harm and violates platform policies. When the clip might seem real, label it as AI in the caption. Many platforms also provide AI-generated content labels.
FAQ
AI Crying trend questions
What is the AI crying trend?+
It's a short-video phenomenon from September 2026. A single photo transforms into a nighttime goodbye scene: you face a soft-focus person, smile through tears, the smile fails, and you walk into darkness, accompanied by a sad ballad. It's based on the CapCut template "Cinematic Emotional Farewell" and gained traction fastest in Indonesia and the Philippines.
How do I make an AI crying video from my photo?+
Press "Create your own" and add one clear face photo. The button shows the cost first. You get the video directly. Farewell is 10 seconds; Rain is 8 seconds. The clip waits in your Library as a 9:16 MP4 with audio. There is no still to approve.
What is the AI crying prompt?+
Two prompts work together: a photo prompt converts your selfie into an over-the-shoulder nighttime close-up with reddened, teary eyes and a delicate smile, and a video prompt animates the smile breaking into tears before cutting to you walking away. Both appear on this page to copy, and the one-tap button executes them automatically.
Is there an AI crying filter I can use?+
Camera app filters typically overlay tears on a live face. This template functions differently: it constructs a complete cinematic scene from your photo and animates it, so the expression evolves and tears move realistically. You receive a shareable file with no watermark, not a filter limited to one app.
Can I add a trending sad song to my AI crying video?+
Yes, in your posting app. The Smew AI file includes original audio, a melancholy piano and strings composition, so there's no copyrighted music. On TikTok, Instagram or YouTube you can layer a trending sound from their licensed collections and lower the original volume.
Can I make an AI crying video of someone else?+
Only of an adult who gave consent and understands your intended use. Never fake a real person's emotional distress, deceive anyone or mock someone without permission. Never use children's photos. Keep it playful.
