Guide to Video Generator AI Models
Discover key differences between models available in Video Generator. Find the best fit for your Text to Video or Image to Video needs in seconds.
Model Availability Overview
Some models are only available for Image to Video, some for Text to Video, and some support both:
Category | Models |
Text to Video only | MiniMax 01 Director |
Image to Video only | Runway Gen-3 Turbo, Runway Gen-4 Turbo, MiniMax 01 Live, MiniMax 01 Subject, Kling 1.6 Pro |
Both | Kling 1.6 Standard, Kling 2.0 Master, Google Veo 2, MiniMax 01 Standard |
Text to Video Models and Parameters
Take a look at how the top Text to Video models compare in terms of supported features:
Aspect Ratio: the shape of the video frame. 16:9 is standard landscape, 9:16 is vertical (for mobile), 1:1 is square.
Duration (5s, 8s, 10s): maximum length of video you can generate in one go.
Quality: maximum output resolution (e.g., 720p = 1280x720px, 1080p = 1920x1080px; exact resolution depends on the chosen aspect ratio).
FPS (Frames Per Second): higher FPS means smoother video (24 is cinematic, 25 is standard in some regions).
Model | 16:9 | 9:16 | 1:1 | 5s | 8s | 10s | FPS |
✓ (1280x720px) | ✓ (720x1280px) | ✓ (960x960px) | ✓ | 24 | |||
✓ (1280x720px) | ✓ (720x1280px) | ✓ (960x960px) | ✓ | 24 | |||
✓ (1280x720px) | ✓ (720x1280px) | ✓ | ✓ | 24 | |||
✓ (1280x720px) | ✓ | 25 | |||||
✓ (1280x720px) | ✓ | 25 |
Still not sure which to choose? Here are some quick tips:
Want longer clips? Google Veo 2 supports up to 8s.
Vertical or square video: only Kling 1.6 Standard and Kling 2.0 Master support all aspect ratios.
Looking for more cinematic control? MiniMax 01 Director is optimized for 15 specific camera motions like [Shake], [Zoom in], [Dolly out], and more
Take a look at our guide to using Text to Video in Video Generator to learn more!
Image to Video Models and Parameters
Image to Video models add extra controls to help you shape the animation’s structure:
First Frame: use your chosen image as the first frame of the generated video.
Last Frame: choose an image that will serve as the last frame of the video.
Subject: upload a reference picture for a character that will be featured in the video (the uploaded image will not be the first frame of the generated clip).
Model | 16:9 | 9:16 | 1:1 | 5s | 8s | 10s | FPS | First Frame | Last Frame |
✓ (1280x720px) | ✓ (720x1280px) | ✓ (960x960px) | ✓ | 30 | ✓ | ||||
✓ (1920x1080px) | ✓ (1080x1920px) | ✓ (1440x1440px) | ✓ | ✓ | 30 | ✓ | ✓ (5s only) | ||
✓ (1280x720px) | ✓ (720x1280px) | ✓ (960x960px) | ✓ | 24 | ✓ | ||||
✓ (1280x768px) | ✓ (768x1280px) | ✓ | ✓ | 24 | ✓ | ✓ | |||
✓ (1280x720px) | ✓ (720x1280px) | ✓ (960x960px) | ✓ | ✓ | 24 | ✓ | |||
✓ (1280x720px) | ✓ (720x1280px) | ✓ | ✓ | 24 | ✓ | ||||
✓ (1280x720px) | ✓ (720x1280px) | ✓ (720x720px) | ✓ | 25 | ✓ | ||||
✓ (1280x720px) | ✓ (720x1280px) | ✓ (720x720px) | ✓ | 25 | ✓ | ||||
✓ (1280x720px) | ✓ (720x1280px) | ✓ (720x720px) | ✓ | 25 | (subject) |
Here's some key details to consider:
Control both first and last frame: only Runway Gen-3 Turbo (all durations) and Kling 1.6 Pro (5s only) support this.
Need 1080p? Only Kling 1.6 Pro offers 1080p for Image to Video.
Longest clips: Runway Gen-3 Turbo, Runway Gen-4 Turbo, and Kling 1.6 Pro (first frame-only) support up to 10s. Google Veo supports 8s clips.
Animating sketches or line art? MiniMax 01 Live is optimized for that use case.
Want to feature a specific character? MiniMax 01 Subject lets you upload a reference image to guide the animation.
Square or vertical video: Kling models, Runway Gen-4 Turbo, and Hailuo MiniMax models support all aspect ratios.
Read our guide to using Image to Video in Video Generator for more helpful tips and tricks.
Whichever model you choose, you’re one prompt away from something incredible!
- Model Availability Overview
- Text to Video Models and Parameters
- Image to Video Models and Parameters