Start + End Frame
Start and end the clip on your own images, by letting the model interpolate the movement between them.
Plan up to six shots, each with its own prompt and references, set keyframes, and generate on MiniMax H3, Kling, Seedance and more.
Bring in your references and split the clip into shots with Multi-Shot.
Multi-Shot
Plan up to six shots in one clip on a timeline, each with its own prompt, length, reference image and snippets. See the tutorial
Text to Video
Describe the action and the camera, and add your @characters, @environment and @motion snippets with @. Discover snippets
text
A Parisian woman is entering a busy Parisian café. As she opens the door, time freezes around her. The people and props in the café that are frozen in time while our character continues to walk. Single shot, dramatic and impactful cinematic time effect. @motion-lateral-track @fx-floating-objectsStart from your images
KeyframesBeta
Set the key moments of your clip with your images, and the model animates everything in between.
References
Load a character, an outfit or a place, and on some models a video or a sound, and refer to each one in the prompt with @ref1, @ref2.
Also in Video
Avatar
Animate a character with a voice or a song. The clip follows the audio and lasts as long as it. Make the audio in Voice or Music.
Motion ControlBeta
Copy the movement of a reference clip, such as a dance, onto your character.
Draft and Finalize
Preview a clip as a cheaper draft, then render the one you keep at full quality. On Flux 3.0 Video, MiniMax H3 and Seedance 2.5.
In the prompt bar
Long-form editing
A full editor for long prompts, not a single line.
Snippet tokens
Add your saved snippets with @. They show as tags in the prompt.
@image references
Refer to each loaded image by its token, such as @image1.
Variables
Give a @variable up to ten values and get one generation for each.
Mode controls
Pick camera, lighting or identity options from menus instead of typing them.
AI Improve
Rewrite your prompt for the selected model, and undo it in one click.
Grammar check
Fix spelling and grammar without changing your tokens and variables.
Translation
Translate your prompt to English, keeping its tokens and variables.
Models in Video
Every model28 models from 9 providers, the same list as the model picker in the app.
MiniMax
- H3
- H3 Max
- H3 Max Turbo
VeoGoogle
- Veo 3.1 Lite
- Veo 3.1 Fast
- Veo 3.1
GeminiGoogle
- Omni 1.1 Flash
SeedanceBytePlus
- Seedance 2.5
- Seedance 2.0
- Seedance 2.0 Fast
- Seedance 2.0 Mini
KlingKlingAI
- Kling 3.0 Turbo
- Kling 3.0
- Kling 3.0 Omni
- Kling Avatar
Luma
- Ray 3.2
GrokxAI
- Imagine Video
- Imagine Video 1.5
- Imagine Video 1.5 Lite
FluxBlack Forest Labs
- Flux 3.0 Video
HappyHorseAlibaba
- HappyHorse 1.1
LTX
- LTX 2.5 Fast
- LTX 2.5 Pro
- LTX 2.3 Fast
- LTX 2.3 Pro
OmniHumanBytePlus
- OmniHuman 1.5
WanAlibaba
- Wan 3.0
- Wan 3.0 Prime
Tutorials
TutorialVideo16 min
Prompt multi-shot videos with a visual editor
Plan a multi-shot video on a visual timeline: one card per shot, each with its own reference image and camera motion.
Questions
How long can a clip be?
It depends on the model. The longest clips, up to 30 seconds, come from Wan 3.0 and Seedance 2.5. With an audio track or a motion clip, the clip takes its length.
Can the clip have sound?
Yes, on many models, with Generate Audio. On a few models the sound is always on.
Can I preview a clip before rendering it at full quality?
On Flux 3.0 Video, MiniMax H3 and Seedance 2.5, yes. A draft is a cheaper preview, and Finalize renders it again at full quality. The result is usually close to the draft, but fine detail or motion can change.
Can I edit or extend a clip I already have?
Yes, in Video Edit: change it from a prompt, continue it, or expand it to a new aspect ratio.
How much does a video cost?
Each model has its own price in credits, listed on the Models page, and you see the cost before you generate.
Ready to create?
Hylope 1.7 is open to everyone. New accounts start with 100 free credits.
