Write a sentence or upload an image and get a video shot in one to three minutes. No camera and no crew. Every leading engine sits on one screen and one balance, and the price is shown before you click.
From text: write a description, get a scene. Fastest, least control over the exact look. From an image: upload an existing image and the engine moves it, which is the best way to control the result. From a first and last frame: two images, and the engine invents the transition, perfect for growth videos. From audio: a song or narration becomes a clip with sound born together with the picture. And video to video: take an existing clip and change something in it, for example re-syncing the lips to a different voice.
No engine wins every category, and anyone saying otherwise is selling something. Kling is the best in the world at holding a real person's face across a video. Seedance gives a long continuous take with sound born together with the picture. LTX-2.5 makes video with sound in one pass, up to 20 seconds, and can also turn an audio file into a clip. And there are Worlds: write a place, go inside, save a camera still, and only then animate it. In one 30-second video it is reasonable to use three different engines, each for the shot it is best at. That is the thing you cannot do when every engine lives on a different site behind its own subscription.
You buy a credit balance and each creation draws a known amount from it. A short low-resolution shot costs little, a long 4K one costs more, and the number is on screen before you click. A month you did not create in costs nothing. The rule that saves the most money: low resolution while you are still searching for the scene, full resolution only once the shot works.
No. The engines understand Hebrew well at the level of intent. Only professional cinematography terms work more precisely in English, because the engine was trained on English descriptions. The fix: write the scene in Hebrew and leave terms like shallow depth of field in English.
Do not ask the engine for it. Video is drawn frame by frame, so letters dance and shift mid-shot, and in Hebrew it is especially bad. Captions go in the edit, where you also get full control over timing and size.
The credits come back to you automatically. You do not have to contact anyone and you do not have to ask.
A world is a 3D place you can walk into, look around, and save a camera still from. 16 credits for the place. The video made from that still is a second charge, on whichever engine you pick. Do not animate the odd 360 thumbnail, only the camera still you saved from inside.
Yes, entirely. For personal, business or commercial use.
Home · Pricing · What's new · All engines · All effects · Help center · Lessons · Meital Sela · AI video · AI images · AI voice · AI music · Image to video · What it costs · Credits, not a subscription · Blog · Connect to Claude and ChatGPT · Presets · Privacy · Terms
Cadabra AI · Meital Sela · Rehovot, Israel