Text to Video
Generate videos from text descriptions with AI models and presets
Create videos from text descriptions - no images or videos required.
AI Models
- Kling: High-quality, versatile
- Hailuo 2.3: Advanced realism
- Seedance: Dynamic, energetic
- Sora 2: Cinematic, professional
- Veo 3: Cutting-edge generation
Presets
Special Effects: Explosions, transformations, weather, magical effects Camera Movements: Zoom, pan, orbit, tracking, crane Crazy Situations: Viral scenarios, unexpected combinations Sora 2/Veo 3 Presets: Movie trailers, advertisements, cinematic styles
How to Use
- Write Prompt: Describe subject, action, environment, camera movement, mood
- Choose Model & Presets: Select model (or use Smart Agent) + apply presets
- Set Parameters: Duration (2-10s), aspect ratio, resolution (
720p/1080p/ higher on supported models — the picker shows tiers the selected model supports), audio (on models with native audio) - Generate & Refine: Review, regenerate with adjustments
Best Practices
Effective Prompts:
- Be specific and detailed
- Describe motion and action
- Include camera movement
- Specify mood and lighting
Model Selection:
- Kling: General purpose
- Hailuo 2.3: Realistic scenes
- Seedance: Dynamic action
- Sora 2/Veo 3: Cinematic quality
Tips for Cinematic Results:
- Use Sora 2/Veo 3 for professional quality
- Apply cinematic presets
- Specify camera movements
- Include lighting and atmosphere details
- Generate multiple variations
- Upscale for final quality
Use Cases
Marketing: Product videos, social ads, brand storytelling Entertainment: Music videos, viral content, memes Business: Presentations, explainers, training Creative: Art projects, storyboarding, concept exploration
Premium models (Sora 2, Veo 3) cost more credits. Resolution tier and native-audio selections multiply the per-second cost on supported models (e.g. 1080p ≈ 1.5× vs 720p on Kling, 2160p up to 4× on Seedance UHD; models with native audio add 1.6–2× when audio is on). Check pricing plan for details.