Fridge POV AI Trend: Star With a Friend in the Tiramisu Fridge Meme
The camera watches from inside the fridge. Your first photo grabs spoonfuls of tiramisu from the tray, the door shuts, then your second photo returns to dig in again. A 10-second clip with spoon and fridge audio. Works with one photo too: try Solo mode, or star your pet.
The fridge POV meme captures a midnight snack from the fridge's perspective: the camera rests on a shelf watching a tiramisu tray, someone opens the door to eat straight from it with a spoon, then returns for seconds. Smew AI turns two photos into a video with two snackers: one who raids first and another who sneaks back.
Also called Fridge cam · Inside the fridge POV · Tiramisu fridge raid
Example
What your Fridge POV video will show
Here's the source reel. Your photos take the place of its people, creating a video with matching shots and your own faces. Each generation produces unique results.
Original reel by @jeanluis_blanco, 10 s
The source reel from @jeanluis_blanco (posted 16 October 2026, approximately 2.7 million likes), labeled AI content on Instagram: filmed from within the fridge, he sneaks tiramisu directly from the tray, shuts the door, then sneaks back for another round. The Two eaters and Solo options recreate this using your photos.
An AI chihuahua take from @barkbuddies01 (posted 18 October 2026): the camera watches as the dog uses a spoon to eat tiramisu, leaves, then returns after the door shuts to lick everything clean. The Pet option recreates this starring your own pet.
Everything happens in your browser. No editing software needed, no prompt required, no watermark added.
Upload your two photos
Tap "Create your own" and upload the first snacker, followed by the second. Sequence matters: whoever you add first appears first on screen. Capture new ones or choose from your gallery.
Tap Generate
The button displays the cost before you start. Seedance 2.5 takes your photos and places them into the Fridge POV trend, recreating it frame by frame for 10 seconds with the original audio. No prompt needed from you.
Save or share it
Your video finishes in up to several minutes as a 9:16 MP4. Save it to your device, post straight from your phone, or retrieve it later from your Library.
What kind of photo works best?
Two eaters needs two photos, one for each person:
The first snacker grabs spoonfuls of tiramisu before the door shuts.
The second snacker returns after the door opens and continues eating from the same tray.
Solo needs one photo of yourself, and Pet needs one of your cat or dog.
Include only one person per shot, with a large, clear face: head and shoulders is plenty.
Face the camera directly, with strong lighting, without sunglasses, masks or hands blocking the mouth.
Only use photos of adults who consented to appear. Never include minors, famous people or strangers.
For Pet: one animal per shot, in focus, showing its face and both ears clearly. Pet photos don't work for Two eaters or Solo, which need a person leaning toward the camera with a spoon.
Where the Fridge POV trend began
Fridge POV shots are an old TikTok joke: the phone sits inside the fridge and films whoever opens it. A "POV: You have siblings" clip filmed from inside a fridge, posted in May 2024, has 54 million views, and a couple's June 2026 video about a camera in the refrigerator, set up to catch a no-bake cheesecake thief, has 1.7 million.
The tiramisu version took off on 16 October 2026, when @jeanluis_blanco posted a reel from inside the fridge with the caption "Quienes se identifican?" ("Who can relate?"). A young man opens the fridge at night, eats tiramisu straight from the aluminum tray, closes the door, comes back, and keeps going until he scrapes the tray clean. Instagram marks the reel as AI content. It passed 2.7 million likes in three days.
The remakes came the next day, and the fridge POV trend became a tiramisu AI trend. On 17 October an AI cat version of the same scene, with a Romanian caption about a mom's tiramisu, passed 1 million views on TikTok, and copies with a woman, a grandpa and a chihuahua followed. Google searches for "tiramisu meme" reached their high for the month on 17 October.
@1.best.memes posts "POV: You have siblings", filmed from inside the fridge; it has 54.2 million views and 7.4 million likes by 19 October 2026. TikTok
@jockanddaja installs a refrigerator camera to catch the no-bake cheesecake thief; it reaches 1.7 million views by 19 October 2026. TikTok
@jeanluis_blanco shares the tiramisu fridge POV reel, "Quienes se identifican?"; by 19 October it collected roughly 2.7 million likes and 19,200 comments. Instagram
@florinsbn5 shares an AI cat remake of the identical fridge moment; by 19 October it reached 1 million views and 40,700 shares. TikTok
@westindianbella uploads her version, "I opened the fridge for ONE BITE"; it collected 306,200 views in the first 14 hours. TikTok
@barkbuddies01 uploads an AI chihuahua sneaking into the fridge to eat tiramisu with a spoon and lick the dish clean; by 19 October it collected 18,500 views. TikTok
Copy and paste
The Fridge POV prompts
You don't need these: the single-tap workflow handles them automatically, along with the trend clip. [Image 1] and [Image 2] are your photos, [Video 1] is the trend clip being recreated.
Two eaters
2 photos: first eater + second eaterSeedance 2.510 s · 720p · 9:16152 credits
Video prompt
Recreate the reference video [Video 1] shot for shot: the same fixed phone camera inside the fridge, on the shelf, looking out at a dark kitchen at night, the same aluminum tray of tiramisu right in front of the lens, the same cold fridge light, the same door moves, cuts and timing. The ONLY change: the man in the video is now TWO different people, one after the other: the person in [Image 1] before the fridge door closes, and the person in [Image 2] after it opens again.
First eater = the person in [Image 1], from the first frame until the door closes: copy the exact face from [Image 1] (the face shape, eye color and eye shape, eyebrows, nose, mouth, skin tone, age, hair color, hair length and hairstyle, and any beard, moustache, glasses or freckles) exactly as in the photo. They wear the everyday clothes from [Image 1].
Second eater = the person in [Image 2], from the moment the door opens again until the last frame: copy the exact face from [Image 2] (the face shape, eye color and eye shape, eyebrows, nose, mouth, skin tone, age, hair color, hair length and hairstyle, and any beard, moustache, glasses or freckles) exactly as in the photo. They wear the everyday clothes from [Image 2].
The same real people from their first frame to their last, so their friends instantly recognize them, also when the face is small in the dark kitchen. Never morph, never drift back to the face of the man in the video, never add or remove facial hair. Do not mix the two faces: [Image 1] is only the first eater, [Image 2] is only the second eater. Both change: no one from the video stays. The first eater never appears after the door closes; the second eater never appears before it opens.
The beats, matching the reference second by second:
0-5 s: the first eater leans into the open fridge and eats big spoonfuls of tiramisu straight from the tray with a metal spoon, chewing with a guilty late-night face, then steps back into the dark kitchen.
5-7 s: the fridge door swings shut and the frame goes black, then the door opens again.
7-10 s: the second eater walks up out of the dark kitchen, leans in, digs a spoon into the same half-eaten tiramisu and eats a big spoonful.
Anatomy rules: one head only, on the person's own shoulders; five fingers on each hand; a normal metal spoon. Exactly one person at a time. Keep the tiramisu tray, the plain milk carton and the real look of a phone video at night. No text, no captions, no logos, no watermark.
1 photo: you, face clearSeedance 2.510 s · 720p · 9:16152 credits
Video prompt
Recreate the reference video [Video 1] shot for shot: the same fixed phone camera inside the fridge, on the shelf, looking out at a dark kitchen at night, the same aluminum tray of tiramisu right in front of the lens, the same cold fridge light, the same door moves, cuts and timing. The ONLY change: the man in the video is now the person in [Image 1], before the fridge door closes and again after it opens.
The eater = the person in [Image 1], in EVERY shot, from the first frame to the last: copy the exact face from [Image 1] (the face shape, eye color and eye shape, eyebrows, nose, mouth, skin tone, age, hair color, hair length and hairstyle, and any beard, moustache, glasses or freckles) exactly as in the photo. They wear the everyday clothes from [Image 1].
The same real person from the first frame to the last, so their friends instantly recognize them, also when the face is small in the dark kitchen and when they come back after the door opens. Never morph, never drift back to the face of the man in the video, never add or remove facial hair. No one from the video stays.
The beats, matching the reference second by second:
0-5 s: the eater leans into the open fridge and eats big spoonfuls of tiramisu straight from the tray with a metal spoon, chewing with a guilty late-night face, then steps back into the dark kitchen.
5-7 s: the fridge door swings shut and the frame goes black, then the door opens again.
7-10 s: the same eater walks back up out of the dark kitchen, leans in, digs the spoon into the same half-eaten tiramisu and eats another big spoonful.
Anatomy rules: one head only, on the person's own shoulders; five fingers on each hand; a normal metal spoon. Exactly one person in every frame. Keep the tiramisu tray, the plain milk carton and the real look of a phone video at night. No text, no captions, no logos, no watermark.
Videos cost credits, with the exact amount displayed on the Generate button before you commit. The single-tap price includes the complete video, created in one step from your photos and the trend clip, billed once. A single video costs 152 credits, so the smallest one-time pack covering it is $39.99 for 450 credits, no subscription required. Lower-cost models are available if you manually run prompts in AI Video.
Video model
Settings
Credits
Seedance 2.5 (this template)
10 s, 720p, recreates the trend clip, one tap
152
One-time packs and plans are on the pricing page. A failed video gives its credits back.
What your fridge POV video shows
Seconds
What happens
0 to 5
Nighttime. The camera watches from the fridge interior, behind a large aluminum tiramisu tray. The first snacker (photo 1) pulls open the door, leans forward and takes multiple spoonfuls straight from the tray, then retreats into the dark kitchen.
5 to 7
The door closes and everything goes dark. Then it opens once more.
7 to 10
The second snacker (photo 2) approaches, leans forward and plunges a spoon into the same partially eaten tray.
This swaps characters in the source fridge POV clip rather than creating new footage: Seedance 2.5 takes the original reel and inserts your two people, keeping the identical fridge lighting, tray and door movements. The result is vertical 9:16, runs 10 seconds, at 720p, ready for Reels and TikTok.
Audio is included: spoon scraping aluminum, chewing sounds, fridge hum and door thumps, without music or dialogue. Tap "Create your own", upload your two photos and tap Generate. The exact cost appears on the Generate button before any charge, a run can take several minutes, and your video appears in your Library once complete.
Solo and Pet: single-photo fridge POV options
Version
Photos
What happens
Two eaters
2
First snacker raids the tiramisu, door shuts, second snacker returns and continues raiding the same tray.
Solo
1
The source reel as originally made: you raid the tray, door closes, you return for seconds.
Pet
1
Your cat or dog rises behind the tiramisu, eats with a spoon, walks away, and after the door shuts returns to lick everything clean.
Solo is the solo midnight fridge run, the 3 a.m. fridge visit nobody confesses to: one person, one tray and zero witnesses. It recreates the same 10-second portion of the source reel as Two eaters, with your face in both segments, making the entire midnight raid about you.
Pet recreates an AI pet remake of the scene: the chihuahua tiramisu video @barkbuddies01 uploaded on 18 October, complete with unique spoon and chewing audio. It appeared a day following the tiramisu cat meme, an AI video of a cat raiding tiramisu straight from its tray, uploaded on 17 October, which exceeded 1 million TikTok views in under two days. Those fridge POV cat and fridge POV dog videos represent the humorous angle of the tiramisu meme, and the Pet option creates one starring your own pet: it preserves its actual face, ears and fur, and paws remain as paws.
All three options of this AI fridge POV run 10 seconds, vertical 9:16 at 720p, with the clip's original audio and no music. The cost appears on each version's Generate button.
Who should raid first, who second?
The humor lands best with two people sharing a fridge and late-night snacks, where both can claim they "only took one bite":
Roommates: the dessert was labeled with a name, yet you both ate it.
Couples: it was reserved for guests, but neither could resist.
Siblings: the timeless fridge raider, the brother or sister who always discovers the tray first.
A parent and adult child: mom prepared it for Sunday, and you both made late-night visits.
Friends: tag the one who constantly says "just one more bite".
It's the who-ate-my-dessert joke captured in one clip. Share it with "Who ate my tiramisu?" and tag your second snacker. The source reel asked "Quienes se identifican?", Spanish for "who can relate?", which captures the essence of every late-night fridge raid: everyone has done this. Your video's tray always shows tiramisu, matching the trend, so viewers recognize the tiramisu fridge meme immediately.
Fridge POV shot, camera in fridge or POV nevera: naming variations
This camera perspective goes by various names. Creators reference it as the fridge POV shot, inside fridge POV shot, camera in fridge shot or camera inside fridge shot, while audiences search POV fridge, POV fridge shot and fridge camera shot. It's the "POV: you are the fridge" perspective from within, the inside fridge view of an opening door, and the POV opening the fridge clip, also searched as POV opening fridge, open fridge POV, POV from fridge, POV from inside the fridge and POV refrigerator.
As a meme format it's the fridge POV meme, POV fridge meme, fridge cam meme, fridge camera meme, midnight snack meme or late night fridge meme, and circulates as a fridge POV GIF as well. The fridge POV TikTok and fridge POV Instagram posts from October 2026 nearly all feature tiramisu, so audiences also search tiramisu fridge POV, fridge POV tiramisu clip, tiramisu trend, tiramisu fridge trend, tiramisu meme TikTok and tiramisu meme fridge clip.
In Spanish, the original's caption makes it the "quien se identifica" clip, with searches for POV nevera, nevera POV, POV desde la nevera, POV refri, POV refrigerador, cámara dentro de la nevera, cámara en la nevera, cámara dentro del refrigerador or tiramisú nevera. Argentina uses heladera for fridge (POV heladera), Brazil uses geladeira (POV geladeira), and French plus Italian posts use frigo (POV frigo).
Best hashtags for posting
Hashtags (TikTok, 19 October 2026): #tiramisu dominates with 788,998 videos and 19.1 billion views, while #latenightsnack holds 236,852 videos and 6.5 billion views. The fridge-specific tags are smaller but targeted: #fridgecam (568 videos, 2.1 million views) and #fridgepov (112 videos, 304,000 views). Spanish posts include #quienseidentifica (40,881 videos). These tags work on Instagram Reels too. Maintain the platform's AI content label.
Caption suggestions: "Who ate my tiramisu?", "It was for the guests", "Just one more bite", "Me and my roommate at 2 a.m.", "Quienes se identifican?". Or write your own message.
Your clip already includes fridge audio, so post it as-is or layer a sound when uploading. This fridge opening trend thrives on TikTok and Instagram, though the fridge trend tag there mainly contains different fridge content (restocks and organization), so fridgecam and fridgepov tags connect with the right audience.
Common issues and fixes
Problem
Fix
Second snacker resembles the first snacker
Use two photos with large, sharp faces looking at the camera, then generate again.
A face resembles the man from the source reel
Use a tighter, sharper photo with good lighting. Small faces in photos produce less accurate results.
The milk carton or a fridge label shifts
It's inherited from the source scene. Generate again if it's distracting.
A strange spoon or unusual hand in one frame
Generate again. Each generation produces new output.
Pet: your pet gets tall upright ears or a tan coat like the chihuahua
Use a sharp photo displaying your pet's face and both ears clearly, then generate again.
Pet: a second dog appears in the dark kitchen, or a paw resembles a hand
Generate again. Each generation produces new output.
The generation fails
Some photos get rejected. Smew AI then attempts a backup model, and if generation still fails, credits are refunded.
FAQ
Fridge POV trend questions
What is the fridge POV meme?+
A video filmed from a fridge's interior: the camera rests on the shelf behind food, and viewers watch someone open the door to eat what they shouldn't, a spin on every meme about nighttime fridge raids. The October 2026 variant shows a tiramisu tray being eaten spoon by spoon at night, with repeat visits, uploaded by @jeanluis_blanco with the caption "Quienes se identifican?".
How do I create the fridge POV AI video?+
Tap "Create your own", upload a clear photo of the first snacker as photo 1 and the second snacker as photo 2, then tap Generate. For Solo or Pet, select that tab and upload one photo. The button displays the cost first. You receive a 10-second vertical video with audio. Generation can take several minutes, and your video appears in your Library when finished. No prompt writing required.
Is there an app for the fridge POV trend?+
No separate app needed. Smew AI operates in your browser on any phone or computer: upload one or two photos and it produces the complete fridge POV clip, ready to share on TikTok or Instagram.
Do I need an actual camera in my fridge?+
No. To film authentically, creators position a wide-lens phone on the fridge shelf behind the dessert and record the door opening. Here the AI produces the entire fridge camera shot from your photos, so you need no camera, no tiramisu and no mess to clean.
Can I create the fridge POV video with my friend?+
Yes, that's the concept: your friend becomes the second snacker, returning for the same tray. Use one clear photo of each person. Both must be adults who consent to participate. Roommates, couples and siblings create the most entertaining pairs.
Can I star as both snackers?+
Yes. Select Solo and upload one photo of yourself: you raid, the door shuts, and you return for more, matching the man in the source reel. On Two eaters you can also upload two different photos of yourself, one in each position.
Which AI model creates the video?+
Seedance 2.5. It takes the fridge POV clip plus your two photos and replaces both people in the clip, a character replacement that preserves the fridge, tray, lighting and door movements. Solo does this with one photo, and Pet replaces your pet into the AI chihuahua remake of the clip. The clip's original audio is restored after generation. If a photo gets rejected, Smew AI attempts a backup model.
What prompt does the fridge POV trend use?+
The creator revealed none: the reel arrived with only a caption. We developed our own prompt matching the style of our other reel recreations. It assigns photo 1 everything before the door shuts and photo 2 everything after it reopens. The chihuahua clip included no prompt either, and on Solo and Pet the single photo covers the entire clip. You don't write any prompt.
How long is the video, and does it include sound?+
It runs 10 seconds, vertical 9:16, at 720p. It includes the clip's audio: spoon scraping the aluminum tray, chewing, fridge hum and door sounds. There's no music or dialogue, so you can layer a song when posting. You can also mute audio on the result screen.
Does the fridge POV video cost credits?+
Yes. The create page displays the exact cost on the Generate button before any charge. If generation fails, credits are refunded.
Will the man from the source reel appear in my video?+
No. The first snacker receives the face from photo 1 and the second snacker receives the face from photo 2. On Solo you perform the entire clip, and on Pet your pet replaces the chihuahua. Smew AI has no connection with @jeanluis_blanco or @barkbuddies01.
Can I feature my pet?+
Yes, select the Pet option and upload one photo of your cat or dog. It recreates an AI chihuahua remake of the fridge clip: your pet stands behind the tiramisu, eats with a spoon, and returns after the door shuts to lick everything clean. Two eaters and Solo are designed for people leaning with a spoon, so pet photos perform poorly there.
Can dogs or cats actually eat tiramisu?+
No, keep actual trays away from them. Tiramisu contains coffee and cocoa, and their caffeine and theobromine are toxic to dogs and cats; many recipes also include raw eggs and alcohol. If your dog ate tiramisu or your cat ate tiramisu, contact your vet. In the Pet video the cat eating tiramisu or dog licking the dish is AI, created from one photo, so no real pet touches a spoon.
Are my photos kept private?+
Yes. Smew AI doesn't post your photos or results anywhere or display them to other users. Videos go to your private Library, and you can delete them anytime.
