AI Models
Video Models
Available AI models for video generation — providers, modes, and capabilities.
Clickraft supports video generation from multiple AI providers, each with different modes, duration options, and output quality.
Providers
Google Veo
Google's video generation models with support for audio generation, high resolution, and multiple generation modes.
| Feature | Details |
|---|---|
| Duration | 4, 6, or 8 seconds |
| Aspect Ratio | 16:9, 9:16 |
| Resolution | 720p HD, 1080p Full HD |
| Audio | Yes — AI-generated soundtrack |
| Reference Images | Up to 3 (Asset mode) or 1 (Style mode) |
| Modes | Text-to-video, Image-to-video, Video Interpolation |
Strengths:
- Built-in audio generation
- 1080p resolution support
- Reference image support for consistent style
- Video interpolation between two keyframes
OpenAI Sora
OpenAI's video models with support for longer durations and video remixing.
| Feature | Details |
|---|---|
| Duration | 5, 8, 10, 15, or 20 seconds |
| Sizes | 1280×720, 720×1280, 1792×1024, 1024×1792 |
| Audio | No |
| Modes | Text-to-video, Image-to-video, Video Remix |
| Models | Sora 2, Sora 2 Pro |
Strengths:
- Longest available durations (up to 20 seconds)
- Video Remix mode for restyling existing videos
- Multiple resolution options including HD sizes
- Strong cinematic quality
fal.ai
A range of video models hosted on fal.ai, including Kling and others.
| Feature | Details |
|---|---|
| Duration | 5, 6, or 10 seconds |
| Aspect Ratio | 16:9, 9:16, 1:1 |
| Audio | No |
| Modes | Text-to-video, Image-to-video |
Strengths:
- Square (1:1) aspect ratio support
- Fast generation times
- Wide selection of specialized models
Mode Support by Provider
| Mode | Veo | Sora | fal.ai |
|---|---|---|---|
| Text to Video | ✓ | ✓ | ✓ |
| Image to Video | ✓ | ✓ | ✓ |
| Video Interpolation | ✓ | — | — |
| Video Remix | — | ✓ | — |
Duration Comparison
| Provider | Min | Max | Options |
|---|---|---|---|
| Veo | 4s | 8s | 4, 6, 8 |
| Sora | 5s | 20s | 5, 8, 10, 15, 20 |
| fal.ai | 5s | 10s | 5, 6, 10 |
Credits
Video generation typically costs more credits than image generation. Credit cost scales with duration and resolution (for Veo). Costs are displayed on the model selector before you run generation.
See Credit System for details.
Tips
- Start short — Use 4–5 second durations to iterate on prompts before generating longer videos
- Use Image-to-Video for more control — providing a starting image produces more predictable results
- Enable Veo audio for videos that need a soundtrack — it's on by default
- Video generation takes longer than images — typical wait is 30–120 seconds depending on duration and model
- Use the first/last frame outputs to chain multiple video clips into a sequence