Video Image Animate SOON Image Edit PRO Video+Audio NEW

AI VIDEO STUDIO

Video + Audio Generation

LTX 2.3 or MiniMax H3 — cinematic video with synced AI audio

Balance: 0.0 tokens

Model

LTX 2.3 — resolution + FPS controls · up to 20s

Concurrent slots 0/3 · 3 free

Reference to Video

Upload 1–5 reference images to lock identity and style. No start-frame or text-only mode in this engine.

Generation Mode

Text to Video

Describe what you want to see

Image to Video PRO

One start frame · animate a still

Batch Input

Template

Loading templates…

Start Image

Pro Feature

Image-to-Video is available to Pro & Ultra subscribers. Upgrade to animate your images.

Upgrade to Pro

Reference Images

Pro Feature

Reference-to-Video is available to Pro & Ultra subscribers. Pro can use 2 refs; Ultra can use 4.

Upgrade to Pro

Reference Images 0 / 5

Select up to 5 — fills empty ref slots

Ref 1 *

Ref 2

Ref 3

Ref 4

Ref 5

Generate audio

Same token cost either way. Name sounds in the prompt when on.

Tip: name specific sounds in your prompt (rainfall, footsteps, dialogue) for better audio.

Reference audio (optional) 0 / 3 · 0s / 15s

Optional voice or sound clip for the model. MP3 / WAV / M4A / AAC / OGG only. Each file max 10 seconds. All clips together must be 15 seconds or less. Not lip-sync. Generate audio (above) still controls whether the clip has a soundtrack.

Audio 1

Tap to add · up to 10s

Audio 2

Tap to add · up to 10s

Audio 3

Tap to add · up to 10s

Audio is a man or woman reference

When on, the prompt is wrapped so dialogue matches the attached voice sample(s).

Prompt expansion

Off by default. Enable only if your prompt is about 2–4 short lines. If your prompt is already detailed, keep this off.

Prompt

Template prompt locked Ultra

This template uses a predefined prompt optimized for best results. Upgrade to Ultra to view and edit the prompt before generating.

Get Ultra to edit prompt

Prompt

Negative Prompt required

Quality

Aspect ratio

MiniMax H3 Size

For best results choose 9 sec and 0.8 megapixels.

Aspect ratio

Megapixels 0.4

0.3 0.4 0.6 0.8 0.9

For R2V at 18–20 seconds, 0.8 and 0.9 MP are unavailable because they can cause out-of-memory errors. The slider is capped at 0.6 MP for this duration.

Resolution

Width 480 px

6440072010401240

Height 832 px

6440072010401240

Aspect ratio: 9:16

Timing

Duration 5 sec

5s10s15s20s

Frame Rate

16 fps 25 fps

Turbo mode

Faster 8-step LoRA (recommended)

People Realism

Natural skin & face detail for R2V · off by default

Estimated Cost

— tokens

Your balance 0.0 tokens