Charme Dance AI: Perform at a Rio Street Party Using Just One Photo
Bright spotlights, hanging bulbs, a historic Rio plaza packed with people after sunset, and there you are, gliding smoothly through charme side steps as the crowd records you. Upload a single photo of yourself, two for a pair, or one of your animal.
The charme dance AI trend places you at a crowded baile charme in Rio after dark: you perform the graceful charme side steps as people film you on their phones, similar to the viral AI grandfather Seu Zé de Madureira. Using a single photo, Smew AI creates your custom 10-second video, either solo, with a partner, or featuring your animal.
Also called Baile Charme AI · Seu Zé de Madureira dance · Pedra do Sal AI dance · Festa Charme AI dance
Example
What the Charme Dance video looks like
This shows the reel that launched the trend. Your clip is a fresh creation built from your photo in this style, not a duplicate of these frames. Each generation produces different results.
Original reel by @seuzerj, 13 s
The AI reel from @seuzerj that launched the trend: white-haired grandfather Seu Zé de Madureira performs charme in the crowded Pedra do Sal scene at night. The embedded text has been cropped out.
The entire process happens in your browser. No editing software required, no prompt to compose, no watermark added.
Upload your photo
Click "Create your own" and upload yourself with a visible face (best from the knees up). Capture it with your camera or select from your library.
Click Generate
The button displays the cost. Wan 3.0 creates a 10-second video directly from your photo and includes the charme track. No prompt writing needed.
Save or distribute
Your clip arrives in minutes as a 9:16 MP4. Save it, distribute from your device, or access it later through your Library.
Which photo works best?
You need one photo for Human, two photos for Duo, and one pet photo for Pet.
Human: a clear image with your face toward the camera, knees up or full body, well-lit. You'll dance wearing the outfit from your photo, so choose something suitable for a street party.
Duo: photo 1 performs on the left, photo 2 on the right. Each photo should contain one person with a clear face.
Avoid sunglasses, masks and beauty filters; ensure no other people appear in human photos.
This template is designed for adults and animals. Use your own image, your animal, or an adult who gave permission. Never use a celebrity or stranger.
Where did the Charme Dance trend start?
Charme is Rio de Janeiro's soul and R&B party dance: small gliding side steps, rolling shoulders and cool turns, danced by whole crowds at the same time. The best-known baile charme meets under the Viaduto de Madureira, and the dance also fills the street parties at Pedra do Sal, a historic square in the old port.
The trend starts with a real clip. On 7 August 2026, the charme dancer @zenegaooficial posted himself dancing in the middle of the Pedra do Sal crowd at night, under bright floodlights: 1.6 million views and 275,100 likes by 18 October. In October, people began to put other faces into that clip with AI. On 16 October, the TikTok account @seuzerj put its white-haired AI grandpa, Seu Zé de Madureira, into it ("nobody believed in grandpa at the baile charme"): the first AI version without a celebrity to take off, with 96,400 views.
On 18 October at 12:12 UTC the first CapCut template tutorials for the charme dance trend came out, and the top one passed 605,000 views in less than a day. Most makers are in Brazil, where creators call it the dança AI retrô of the Festa Charme.
Charme dancer @zenegaooficial shares his performance in the crowded Pedra do Sal scene after dark: reaching 1.6 million views and 275,100 likes by 18 October. TikTok
He performs there again with a companion, side by side: collecting 470,400 views. Duo charme videos become part of the trend. TikTok
Initial AI modifications of the Pedra do Sal footage emerge: Hollywood action figures perform in the crowd. They remain minor (2,264 views for the earliest one). TikTok
@seuzerj shares the AI grandfather Seu Zé de Madureira performing charme in the identical crowd: reaching 96,400 views, the first AI adaptation without a famous person to gain popularity. TikTok
The first CapCut template tutorials come out at 12:12 UTC; by the next day one has 605,800 views and four more have 119,800 to 323,400. TikTok
English guides begin appearing. #charmedance shows 4,701 TikTok posts and 2.9 million views, #bailecharme displays 14,533 posts and 182 million views.
Copy and paste
The Charme Dance prompts
You won't need this: the single-tap process handles it automatically. [Image 1] represents your photo: the video system uses it as reference and creates the scene directly from it, skipping the intermediate photo stage.
Human
1 photo: you, face clear (knees up is best)Seedance 2.5 Video Edit13 s · 720p · 18:23126 credits
Video prompt
Edit [Video 1]. Replace only the explicitly mapped source subjects in place:
[Image 1] replaces the foreground dancer with the large white afro, glasses and red polo shirt, tracked through the front and profile views, 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: Human / main. 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.
Replace only this foreground dancer with the uploaded adult identity. Keep the separate dancer in the hat and white sleeveless shirt, crowd, street, original red polo outfit, handheld camera and the exact source dance rhythm. Do not put the uploaded face on the partner or any onlooker.
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.
2 photos: you (left), face clear + your friend (right), face clearSeedance 2.5 Video Edit13 s · 720p · 18:23126 credits
Video prompt
Edit [Video 1]. Replace only the explicitly mapped source subjects in place:
[Image 1] replaces the foreground dancer with the large white afro and red polo shirt; photo slot: You (left), face clear, in every shot where that subject appears.
[Image 2] replaces the separate dancer in the hat and white sleeveless shirt; photo slot: Your friend (right), face clear, 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.
[Image 2]: 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: Duo / duo. 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 Charme Dance / Duo 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 pet, whole bodySeedance 2.5 Video Edit13 s · 720p · 18:2332 credits
Video prompt
Edit [Video 1]. Replace only the explicitly mapped source subjects in place:
[Image 1] replaces the foreground dancer with the large white afro and red polo shirt; photo slot: Your pet, whole body, in every shot where that subject appears.
[Image 1]: Replace the ENTIRE mapped performer in EVERY source shot with the uploaded pet's actual species, face, muzzle, eyes, ears, fur, markings and coherent animal body. Replace human skin, hands, fingers, torso, sleeves and clothing belonging to this target with the pet's own fur and natural paws. Foreground hands belonging to this performer are included. Paws follow source gesture timing and limb trajectories without becoming five-fingered human hands. Never put a pet head on the source human body. Other separately mapped human roles stay human.
Selected variant: Pet / pet. 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 Charme Dance / Pet 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
Human
Seedance 2.5 Video Edit
13 s · 720p · 18:23
126
Duo
Seedance 2.5 Video Edit
13 s · 720p · 18:23
126
Pet: Standard
Seedance 2.5 Video Edit
13 s · 720p · 18:23
32
Pet: Best
Seedance 2.5 Video Edit
13 s · 720p · 18:23
126
One-time packs and plans are on the pricing page. A failed video gives its credits back.
What happens in the charme dance AI video, second by second?
Each variant presents one uninterrupted handheld phone recording from within the crowd, lasting 10 seconds, vertical 9:16:
Time
Human and Duo
0-3 s
From knees up surrounded by people: you move to the rhythm and begin the charme movement, sliding side to side.
3-6 s
Two smooth movements to one direction and return, one leg crossed behind, a stylish glance sideways. The crowd reacts.
6-8 s
You move toward the phone with a wide smile, your face prominent and focused.
8-10 s
Return to the crowd, continuing to dance, as people record you.
The setting remains consistent across all three: a historic Rio plaza after dark, painted colonial buildings, hanging lights across the street, white spotlights and a crowd of celebrants with phones raised. Human, Duo and Pet use Wan 3.0 directly from your photos. Each video includes the charme track and crowd audio. Each generation produces a fresh interpretation, so outcomes vary.
What is charme, and what is a baile charme?
Dança charme, o que é? Charme represents the dance style of Rio's Black music celebrations since the 1980s: soul, R&B and slow funk, performed with graceful, synchronized movements rather than jumping. A baile charme is the celebration itself; baile charme, o que é in brief: an evening where everyone performs the same flowing moves to classic R&B. People learning it search for dança charme passo a passo guides, with traditional songs known as charme das antigas.
The most celebrated baile charme, the Baile Charme do Viaduto de Madureira, occurs beneath a highway overpass in Madureira, explaining why searches for baile charme Madureira and the Viaduto de Madureira connect with this trend. The footage the trend replicates was recorded at the Pedra do Sal street celebration in downtown Rio. This explains why people also refer to it as the Rio street party AI trend, or the charme dance Brazil trend.
Who is Seu Zé de Madureira, and is he real?
Seu Zé de Madureira serves as the icon of the Seu Zé de Madureira trend: a silver-haired grandfather wearing round spectacles, a red shirt and suspenders who performs charme more skillfully than the crowd. He represents an AI persona of the TikTok account @seuzerj, which inserts him into viral clips from other creators and acknowledges the source in each description. His baile charme video from 16 October takes @zenegaooficial's Pedra do Sal performance and replaces the dancer with the grandfather: identical crowd, identical spotlights, identical choreography. Therefore the Seu Zé de Madureira IA footage does not document an actual party attendance.
This page presents that footage as the source AI reel, with attribution. Our template doesn't replicate his appearance or reference his name in any prompt: it features you, your companion or your animal at the celebration, maintaining your authentic face and attire.
What goes wrong, and how do you fix it?
Problem
Fix
The face appears slightly distorted during a head rotation
Choose a focused photo with the face directed at the camera, well-illuminated, and generate again.
The two faces merge on Duo
Choose two photos with one individual each; maintain photo 1 for the left performer and photo 2 for the right.
The dance appears rigid
Charme is a relaxed, graceful dance, so movements are deliberately subtle. Generate again for a different interpretation.
Your animal returns to four legs
Select Best for the animal, or choose a photo showing the complete animal.
A brand mark or team jersey appears in the crowd
The prompt excludes brand marks and team jerseys. Generate again if one appears.
The video system rejects the photo
Certain photos get rejected. The credits are refunded; attempt a different photo.
FAQ
Charme Dance trend questions
What is the charme dance AI trend?+
It represents the charme dance trend from October 2026: AI inserts an individual into a crowded Rio street celebration after dark, where they perform the graceful charme side steps as the crowd records them. It originated from an authentic Pedra do Sal dance footage from 7 August 2026 and gained popularity with the AI grandfather Seu Zé de Madureira on 16 October. Smew AI generates your personal adaptation from a single photo.
How do I make a charme dance AI video?+
Click "Create your own", select Human or Duo, and upload your photo (two for Duo). Click Generate; the button reveals the cost before any charges. You receive a 10-second vertical AI dance clip with its unique rhythm and crowd audio. No prompt composition required.
Do I need the charme dance CapCut template?+
No. The charme dance CapCut template is a restricted template, with the charme dance tutorial clips on TikTok explaining how to access it in CapCut. Smew AI operates in the browser as an AI dance video creator: no CapCut, no application, no editing work.
Can I dance charme with a friend?+
Yes. Select Duo and upload two photos: photo 1 performs on the left, photo 2 on the right, both synchronized, in the identical Rio crowd. The source dancer shares duo videos with companions as well, making the charme dance duo variant faithful to the original.
Can my dog dance charme?+
Select Pet. It shows an AI dog performance at the baile: your dog (or cat) rises on back legs and executes the charme side movements in the crowd. This dog dancing AI footage represents the charme dance pet variant, compatible with any animal in one clear photo.
Is Seu Zé de Madureira a real person?+
No. Seu Zé de Madureira is an AI grandfather persona from TikTok account @seuzerj. His charme footage is an AI modification of @zenegaooficial's authentic Pedra do Sal dance recording. Your Smew AI footage is also AI-generated, so preserve the platform's AI indicator when posting.
What music is in the video?+
Each video includes an authentic smooth soul and R&B rhythm with crowd audio, created by the video system for your footage. It isn't the track from the viral videos: that track has its own copyright holders. You can also include a sound of your preference when posting.
What prompt does the charme dance AI trend use?+
No creator revealed one: the CapCut template is restricted and the AI videos don't disclose prompts. Our prompts are original, developed from the source footage (the historic plaza, the spotlights, the crowd, the choreography) and a brief trend summary: dance charme with graceful side movements in a packed evening street celebration in Rio, hanging lights, people recording with phones, a handheld phone recording.
Which AI models make the videos?+
Human, Duo and Pet all use Wan 3.0 directly from your photos, at 720p. Processing may require several minutes.
Como fazer o vídeo da dança charme com IA?+
Envie uma foto (ou duas, para dançar em dupla), escolha Human, Duo ou Pet e toque em Generate. O vídeo de 10 segundos mostra você dançando charme no meio de um baile charme lotado no Rio, à noite. É a mesma ideia do baile charme IA e da dança AI retrô do TikTok, sem CapCut e sem app: a dança charme IA feita direto no navegador, com uma foto.
Is there an app for the charme dance AI?+
No application required. People search app dança ia, dança ia foto or video charme ai because the trend originated with an app template; Smew AI produces the identical charme AI footage on the web from your photo, as one of our AI dance trend templates, functioning like a make my photo dance feature for this specific celebration.
Is it funk or charme?+
Both terms appear. Many guides label it Brazilian funk dance AI, funk dance AI or ia dança funk, because the template's original description identifies it as a Brazilian funk dance. The choreography in the footage is charme: slow, graceful soul movements, not the rapid funk carioca style. Our footage maintains the charme movements.
How long is the video and what format is it?+
All variants produce 10 seconds, vertical 9:16 MP4 with audio and no watermark. Human and Duo render at 720p; Pet Standard at 1080p and Pet Best at 720p.
