Skip to main content
Text to Video lets you generate cinematic video content from a written description alone. Describe a scene, an action, or an environment and the AI produces a dynamic video to match — no footage, cameras, or editing software required.

How to generate a video

1

Open Video mode

Navigate to Video mode from the left sidebar. This opens the Video Studio interface where you enter your prompt and configure settings.
2

Write your prompt

Type a detailed description of the video you want to create in the prompt field. Be specific about the scene, action, characters, camera movement, and visual style. The more precise your description, the closer the output will be to your vision.Example prompts:
  • A lone astronaut walks across a barren red desert at dusk, golden light casting long shadows, cinematic wide shot
  • Close-up of a raindrop hitting a still puddle in slow motion, water ripples spreading outward, hyper-realistic
  • A futuristic city skyline at night with neon signs reflecting on wet streets, flying cars weaving between towers
  • A majestic eagle soaring over snow-capped mountain peaks, aerial tracking shot, natural lighting
Include camera directions (e.g., “tracking shot,” “slow zoom,” “overhead view”) and lighting descriptors (e.g., “golden hour,” “soft studio light”) to improve consistency with your creative intent.
3

Select a video model

Choose the AI model that best fits your style and requirements. Each model has distinct characteristics:
ModelBest for
Sora 2High-fidelity cinematic sequences
Google Veo 3Photorealistic scenes with natural motion
Kling 3.0 ProFlexible durations up to 15 seconds
Hailuo O2 ProHigh-quality 1080p output
WAN 2.5Fast, cost-effective generations
Seedance 1.5 ProWide range of aspect ratios and durations
Runway 4.5Versatile formats including 4:3 and 3:4
Hover over any model name in Video mode for approximately 3 seconds to see its credit cost before committing to a generation.
4

Configure aspect ratio and duration

Select the aspect ratio and duration that suit your project. Available options vary by model.Aspect ratio guide:
RatioBest use
16:9YouTube, presentations, standard video
9:16Instagram Stories, TikTok, mobile-first
1:1Instagram feed posts, social media
4:3Traditional TV format, PowerPoint
3:4Posters, digital flyers, tablet display
Duration: Most models support 5 or 10-second clips. Kling 3.0 Pro supports durations from 3 to 15 seconds in 1-second increments.
Longer durations and higher resolutions consume more credits. See Video Credits for a full breakdown.
5

Generate your video

Click Generate. The AI processes your prompt and settings and produces the video, typically within 30–60 seconds depending on the model, duration, and current queue.Once complete, the video appears in your creations. From there you can download it, extend it with Video Extend, apply Video Effects, or use Lipsync to add a talking avatar.

Supported models

The following models support Text to Video generation:
ModelAspect ratiosDuration (seconds)
Seedance Lite4:3, 16:9, 1:13–13
Imagine V39:16, 16:9, 1:15
Kling 1.6 SD9:16, 16:9, 1:15, 10
Kling 1.6 Pro9:16, 16:9, 1:15, 10
Kling 2.1 SD9:16, 16:9, 1:15, 10
Kling 2.1 Pro9:16, 16:9, 1:15, 10
Kling 3.0 Pro16:9, 9:16, 1:13–15
Hailuo AI16:96, 10
Hailuo O2 SD9:16, 16:9, 1:1, 2:3, 3:2, 3:4, 4:36, 10
Hailuo O2 Pro9:16, 16:9, 1:1, 2:3, 3:2, 3:4, 4:36, 10
LUMA RAY29:16, 16:9, 3:4, 4:35
PixVerse 3.59:16, 16:9, 1:1, 3:4, 4:35, 8
PixVerse 49:16, 16:9, 1:1, 3:4, 4:35, 8
PixVerse 4.59:16, 16:9, 1:1, 3:4, 4:35, 8
PixVerse v59:16, 16:9, 1:1, 3:4, 4:35, 8
Google Veo 29:16, 16:95
Google Veo 39:16, 16:9, 1:15
Google Veo 3 Fast9:16, 16:9, 1:15
Seedance Pro1:1, 4:3, 16:9, 3:4, 9:165
Seedance 1.5 Pro6:9, 9:16, 3:4, 4:3, 1:14–12
WAN 2.29:16, 16:9, 1:15
WAN 2.59:16, 16:9, 1:15
Sora 29:16, 16:95, 10
Lucy9:16, 16:95
Runway 4.516:9, 9:16, 1:1, 4:3, 3:45, 8, 10

What to do next

Image to Video

Use a static image as the starting frame of your video.

Video Effects

Apply VFX effects to your generated video.

Extend Video

Add 5 more seconds to an existing generation.

Video Credits

Understand how credits are calculated per model and setting.