Video
Free AI video generator in Telegram
A free AI video generator that lives in Telegram: describe a scene, or send a photo and watch it move. Nothing to install, no sign-up, and the clip comes back to the chat you started from.
Make a video12 Stars per week free, no card. The cheapest video model starts at 4 Stars, so one clip a week costs nothing, and unused Stars do not carry over to the next week. What the free allowance covers
Catalogue
Models for video
Everything running today, with what the cheapest run of each model costs. A subscription never unlocks or locks one of them: you pick.
15 modelsfrom 4 Stars
Seedance 2.0
ByteDance's flagship: it holds a character and a scene from shot to shot instead of rebuilding them every second. It comes in 720p, 1080p, and an edit pass over video you already have.
from 36 Stars
Pruna P-Video
The cheapest video here. The draft rung costs less than the free weekly allowance, which means your first video need not cost you anything at all.
from 4 Stars
Wan 3.0
First in the world on blind text-to-video comparisons, and a third the price of the flagship we used to lead with.
from 32 Stars
Veo 3.1 Fast
Google's video model with sound, and the sound is real speech rather than dubbed-in effects. The only model here that does it.
from 41 Stars
Hailuo 2.3 Fast
The quick branch of Hailuo: it turns a photo into six seconds of movement, and the prompt is optional. It needs an actual photo, text alone will not start it.
from 25 Stars
Kling 3 Pro
Kling 3 Pro writes the sound along with the clip: footsteps, room tone, atmosphere. It starts from a photo and gives the cleanest motion in the Kling line.
from 67 Stars
Grok Imagine Video
xAI's video model: a photo becomes a five-second scene, and a clip you already have becomes a new version of itself. What you attach decides which of the two you get.
from 45 Stars
Seedance 2.0 Mini
The small Seedance: the same feel for movement as the flagship, in a single 720p rung driven by text. For drafts and for testing an idea before you run the bigger model.
from 57 Stars
Hailuo 2.3
The same Hailuo, but it starts from text: describe the scene and get six seconds. It keeps the physics honest, so people and objects move the way you expect.
from 36 Stars
PixVerse V6
Inexpensive video with sound. The middle ground for when Veo is too dear and a draft is not enough.
from 23 Stars
Kling 2.1 Standard
Classic Kling: it turns a photo into a five-second clip with a smooth camera move. The workhorse for when you want movement without surprises.
from 34 Stars
Pruna Avatar
One photo plus a script, and the person in the photo says it. Thirty voices to choose from.
from 32 Stars
Veo 3.1
Google's video model for motion and sound. Describe a scene or start from a photo, then choose length, resolution, and audio.
from 101 Stars
Runway Gen-4.5
Runway Gen-4.5 makes video from a description or a starting image. Choose the frame shape and a length from 2 to 10 seconds.
from 31 Stars
Luma Ray 3.2
Luma Ray 3.2 creates video from text or animates a photo. Choose from three resolutions, two text-video lengths, and a seamless loop for five-second clips.
from 19 Stars
How to choose
Which model to pick
A shelf this wide only helps once you know which one to open first. Here is the short version, sorted by what you are actually trying to do.
Trying an idea out
Start on Pruna P-Video. Its draft rung is the cheapest run on this page, and a draft answers the question worth answering early: is this scene worth paying a flagship for?
4 Stars is where video starts here, under the free weekly grant, so the first clip need not cost you anything. Hailuo 2.3 Fast is the cheap step up once a draft looks right.
When it has to look finished
Wan 3.0 came first in blind text-to-video comparisons and still sits under the old flagship prices. Seedance 2.0 trades differently: it holds a character and a set across shots instead of rebuilding them every second.
Veo 3.1 Fast is the only model here that writes real speech into the clip, so take it when the video has to be heard and not only watched. Kling 3 Pro and PixVerse V6 come back with sound too, at very different prices.
When you already have the photo
Animating a frame is its own mode, not a checkbox. Hailuo 2.3 Fast and Kling 2.1 Standard both take a photo and move it, and the prompt is optional: they will find the movement themselves.
Pruna Avatar is the outlier: a photo plus a script, and the person in the photo says it. Send the shot straight from the chat, there is nothing to upload.
What Telegram changes
Being a free AI video generator in Telegram removes the download step: the finished clip lands back in the chat you started from, ready to forward to a person or a channel.
12 Stars arrive every week with no card attached, and they do not roll over: a new week replaces what is left rather than adding to it. Everything past that is paid in Telegram Stars, one credit per Star, and a failure on the provider's side refunds itself.
Process
Three steps to a finished clip
- Open the bot@StarGenAI_Bot, and the Mini App from there. Nothing to download.
- Describe the scene or send a photoText, a frame, or both: a photo turns into a moving shot.
- Take the clip back to the chatThe video lands in the same chat you started from. Forward it wherever you like.
FAQ
Questions about video
Is the AI video generator really free?
You get 12 Stars every week and no card is needed. The cheapest video model starts at 4 Stars, so one clip a week costs nothing. Unused Stars do not carry over: the new week replaces what is left instead of adding to it.
What does a clip cost?
Every model has its own price, listed above. The same amount sits on the generate button before you start, and that is exactly what is charged: one credit is one Telegram Star.
Can I animate my own photo?
Yes, that is its own mode. Send the frame, describe the movement, and the model builds a clip out of it. Nothing to upload: the photo is already in Telegram.
What if the clip comes out wrong?
When a generation fails on the provider's side the credits come back on their own, so you never pay for someone else's error. If you simply did not like the result, change the description and run it again.