The Im Not Your B No More AI dance is a 13-second video in which you take the place of a performer on a gym stage, bouncing on bent knees, spinning and throwing an arm out while a crowd holds up phone lights. On videosi you make it from one photo, with no prompt, and the trend sound comes already on the clip.
This guide answers the questions the Im Not Your B No More AI page does not cover: where the clip comes from, what each beat does, what the AI changes, which photo works and what to check before posting. It is independent of the creators involved.
Key takeaways
- The clip runs 13 seconds in six beats: arm up, knee bounce, spin, arm sweep, crowd pan and a finish on stage.
- The first upload is unconfirmed; the earliest clip we could check is a repost dated 8 October 2026.
- Only the dancer is replaced. The stage, crowd, camera and timing stay as in the original.
- A full-body, evenly lit photo of one person gives the best likeness.
- A 13-second Standard 720p video is 11 credits; label the post as AI and expect platforms may mute the sound.
In this guide
- Where does the dance come from, and how sure are we?
- What happens in the 13 seconds, beat by beat?
- Is there a fixed routine to learn?
- Why do people spell the caption so many ways?
- What does the AI actually change in the clip?
- Which photo gives the best result?
- Why can the face drift after the spin?
- What does a try cost in credits?
- How does the trend sound end up on the video?
- What should I check before posting?
- Why not just prompt a general model yourself?
Where does the dance come from, and how sure are we?
The dance is built on a clip of a performer on a low wooden stage in a gym hall. The earliest clip we could check is a repost dated 8 October 2026. That repost carries a watermark from another account whose posts need a login to view, so the first upload and its creator are unconfirmed.
That caveat matters if you want to credit someone. Any source that names a single originator for Im Not Your B No More is going further than the evidence we found. The radar behind videosi first listed the trend on 30 October 2026, as a budding trend, and the videosi page is dated 1 November 2026.
What happens in the 13 seconds, beat by beat?
The clip has six beats. In the first second the performer is mid-move at the edge of the frame with one arm high. From 1 to 4.5 seconds he bounces on bent knees with a microphone hand raised while a handheld camera settles on him.
From 4.5 to 6.5 seconds he turns his back to the lens and spins, hair and hoodie flying. Between 6.5 and 7 seconds one arm cuts out toward the room and the camera follows it. From 7 to 9.5 seconds the shot pans across a packed hall with phone lights up and the dancer is out of frame. The last 3.5 seconds return to the stage for more knee bounces and a long arm thrown out to finish.
Is there a fixed routine to learn?
No. Beyond the bounce, the spin and the arm, the dance has no single fixed routine. That is part of why it spreads: anyone can do a version of it in a bedroom, an office or a street.
Remakes on the main sound usually tag it GO. The AI version does not teach a routine at all. It copies the motion and timing of the original clip and puts you in the performer's place, so you do not need to know any steps.
Why do people spell the caption so many ways?
The name comes from how people caption the clip. The line is censored, and posts spell it many ways, such as not yo b, not your bih and not his b. People read it as a declaration of being done with somebody, a way of saying you are no longer anyone's partner or pushover.
An older, separate run of not your b no more posts from earlier in 2026 uses different sounds, and those are not this dance. If you search for the trend, add the dance or the stage detail so you land on the right posts. On videosi the text is never shown on the video.
What does the AI actually change in the clip?
Only the dancer. Your face, hair and outfit come from your photo, while the stage, the crowd, the camera moves and the timing come from the original clip. His cap and hoodie are named as gone, so they should not carry over.
Behind the scenes the video is made with Seedance, ByteDance's video model, from a photo plus a motion reference. The prompt tells it to replace the one named person on stage and to keep the room. Nobody in the crowd borrows your face, and there are no logos, text or lip sync. We do not publish the prompt itself.
Which photo gives the best result?
Use a full-body photo of one person in even light. The dance is all legs and arms, and a head-and-shoulders crop leaves the model to invent your trousers and shoes. Your clothes and hair in the photo are what you wear on stage.
Wear something that moves, such as a jacket or hoodie, because it shows off the spin better than a tight top. Keep your face clear of sunglasses and low hats. The camera sits at the side of the stage, so your face needs to be sharp to survive it.
Why can the face drift after the spin?
The spin hides the face for about two seconds, and likeness can drift when it comes back. For much of the clip your face is also seen from the side, which is how the original was filmed. These are limits of the source footage rather than a setting you can change.
Likeness is not guaranteed. A sharp, bright photo gives the model the most to hold on to. If a render fails, times out or is blocked, its credits come back automatically.
What does a try cost in credits?
There is no free generation. A 13-second Standard video at 720p is 11 credits, and an account's first video on that tier costs 10. Standard at 480p is 6 credits, Fast at 720p is 19, Pro at 720p is 22 and Ultra at 720p is 41.
Credit packs never expire. The Single pack is $6.99 for 10 credits, the Trio Pack is $17.99 for 30 and the Duo Pack is $29 for 50. The Trend Pass is $9.99 a month for 30 credits, and the Basic plan is $29.90 a month for 100. Prices can change, so check the page before you buy.
How does the trend sound end up on the video?
The model renders the picture silently, and the trend sound is added afterwards, aligned to the clip. You download an MP4 with no watermark and the sound already on it, so there is nothing to line up yourself.
Most remakes checked use a TikTok sound labelled "original sound - steveglover69", but no official song title or performer is confirmed. Creators also use other uploads of it.
What should I check before posting?
The sound belongs to its rights holders and the line is explicit in its uncensored form, so a platform may mute or limit the post. Turn on the platform's AI-generated label, and check the current rules of the app.
Only use a photo of an adult who agreed to it. The original crowd, the photographer and the hall stay as they are, and faces there are not yours. The mouth does not sing along, so keep captions and expectations realistic. You can make the video again if the first one is off.
Why not just prompt a general model yourself?
You can, but a general model treats the reference as footage to edit, so a loose prompt often leaves the original performer in place. The cap and long hair are strong features and tend to survive unless they are named as removed.
Each try also costs model credits and several minutes, and the sound still has to be lined up afterwards. The Im Not Your B No More AI page builds the prompt, the reference and the sound in, so you only choose a photo and a format.
To try it, open the Im Not Your B No More AI page, upload one photo and press Generate.
