Introduction
Gen-3 Alpha is the first of upcoming models that offers improvement in fidelity, consistency, motion and speed over previous generations of models. Gen-3 Alpha is currently available to users on a Standard plan or higher.
Gen-3 Alpha Turbo is a faster model in the Gen-3 Alpha family that generates at a lower cost. The Turbo model is available on all plan levels and requires an input image.
This article outlines the steps to create videos with Gen-3 Alpha, the available settings and more.
Article highlights
- The Turbo model requires an input image, so switch to Gen-3 Alpha for text-only prompting
- Use a highly descriptive prompt when using Text to Video on the Gen-3 Alpha model
- Focus on describing the desired motion when using an input image
- A single generation can be extended up to three times
Related Links
Spec Information
| Spec | Gen-3 Alpha | Gen-3 Alpha Turbo |
|---|---|---|
| Cost | 10 credits per second | 5 credits per second |
| Supported durations | 5 seconds 10 seconds |
|
| Explore Mode on Unlimited Plans | Yes | |
| Platform availability | Web, iOS | |
| Base prompt inputs | Text Image |
Text Image (Required) |
| Text character limit | 1000 characters | |
| Output resolutions | 1280x768 | 1280x768 768x1280 |
| Keyframes support | First or last frame | First, middle, and last |
| Video Extension increments | 5 or 10 seconds | 8 seconds |
| Maximum extended length | 40 seconds | 34 seconds |
| Frame Rate (FPS) | 24fps | |
Step 1 – Drafting the Prompt
Begin by navigating to Generative Session in your Dashboard.
From here, make sure that Gen-3 Alpha or Gen-3 Alpha Turbo is selected from the dropdown in the bottom left corner.
To use Text to Video, please ensure that you select the Gen-3 Alpha model. The Turbo model requires an input image.
Text Prompts
Gen-3 Alpha can create highly detailed videos with complex scene changes, a wide range of cinematic choices, and detailed art directions. A descriptive yet clear prompt is key to generating a great video.
Add a descriptive text prompt that conveys the camera angle, subject, scene, style and movement to generate your video. Check out our Gen-3 Alpha Prompting Guide for ideas and more prompt examples.
Below are text-only prompt examples along with their respective outputs:
| Prompt | Output |
| A dramatic zoom in on the face of movie villain as he raises an eye brow and the lights shift, casting an eerie red glow across him. Evil villain lair, 1980s spy movie, cinematic, 35mm film, dynamic movement. | |
| A sci-fi-like action chase scene, FPV hyper-speed fly through multiple locations. Racing through asteroid fields, through a dense clouds, through a complex system of desolate landscapes. Dynamic motion, dynamic blur, timelapse, 30x speed, cinematic, muted color palette. | |
| Dynamic motion, 30x speed. Camera follows a translucent white plastic grocery bag with bold red letters printed on it that read "THANK YOU" as it flies organically in the wind of a desert. the slightly opaque bag undulates in the wind, maintaining the bold red "THANK YOU" text printed on it. |
Image and Text Prompts
Input images are optional in Gen-3 Alpha, but required in Gen-3 Alpha Turbo. Input images will act as the first frame of your video by default.
You’ll be prompted to crop your input image if it is not in a supported resolution.
Include a simplistic text prompt to guide the output of your video. Instead of describing what is in the image, focus on describing the movement of the camera, character, and scene you'd like in the output.
Describing the full contents of input images may lead to unexpected results.
Below are examples of input images, text prompts that focus on motion, and their respective outputs:
| Input image | Prompt | Output |
| the gloved hands pull to stretch the face made of a bubblegum material |
|
|
| the sea anemones sway and flow naturally in the water. the camera remains still. | ||
| subject stiffly walks, his movement hindered by the heavy armor. dynamic motion. camera zooms out to retain framing as he moves closer. |
Step 2 – Configuring the Settings
Gen-3 Alpha has a few additional settings that you should review before starting your generation.
Keyframes
You can choose if you'd like your input image to act as the first or last frame in Gen-3 Alpha, or configure both the first, middle, and last frame on Turbo. Please see Creating with Keyframes for more information on using this feature.
Camera Control
Use Camera Control to choose both the direction and intensity of how you move through your scenes for even more intention in every shot. Please see Creating with Camera Control for more information on using these settings.
You can configure the following by selecting the Settings Icon in the bottom left hand corner:
Fixed seed
Using a Fixed Seed will allow you to create similar generations. This is unchecked by default to give you a wide variety of results.
Copy and paste the seed of a previous output if you'd like to receive generations with similar style and movement. Pasting the seed will automatically check the box.
Aspect ratio
On the Turbo model, you can choose between 1280x768 and 768x1280 aspect ratios before starting your generation. Changing this setting may prompt you to crop any currently selected input images.
Step 3 – Generating the Video
After drafting your text prompt and configuring your settings, you're now ready to generate your video.
You can choose between a 5 or 10 second duration for your output with the duration dropdown near the Generate button. Generative Video defaults to 10 second generations.
Your generations will be scrollable through your session as you continue to generate. You can also access completed videos in your Assets, where they will save to the Generative Video folder by default.
Once your generation is complete, you'll notice a few additional options at the top right of the video.
Clicking the arrows will allow you to reuse the settings used to generate the video:
You'll also see an Actions button under the video:
Clicking this button will expand more options to continue working with this video:
Extend
Use this option to extend the duration of your video by processing another generation. Learn more in Step 4 of this article.
Lip Sync
Use this option to use the output with Lip Sync.
Video to Video
Select this option to process the generation in Video to Video.
Edit Video
Clicking this button will allow you to edit the duration, speed, and handheld camera shake of your generated video, as well as reverse it. After adjusting the settings to your liking, click Render to process the video.
Trim
To trim your video, drag either end of the timeline below the generated clip inwards. You will see the new duration of the clip update automatically as you drag.
Retime
To retime your video, move the Playback speed slider. Moving the slider to the left will slow the clip down; moving it to the right will speed it up. The default setting is 100%, which is your clip’s original speed.
Handheld shake
Use this option to add a handheld camera shake effect to your video. Shake strength will determine the intensity of the shake, and Shake speed will determine how fast the shake is.
Reverse
To reverse your video, toggle on the Reverse button.
Expand Video
Reframe the video to a different aspect ratio by processing another generation. Learn more in Creating with Expand Video.
Upscale Video
Re-generate the clip in a higher resolution by selecting Upscale to 4k.
Please note that once a video has been upscaled to 4k, Reuse Settings is no longer available.
Download options
You can download the generation directly from your Session by clicking on the download button at the top right of your generated clip. Here, you can choose to download your clip either as an MP4 or a GIF.
Step 4 – Extending the Video
Completed Gen-3 Alpha and Turbo generations can be extended up to three times to create a longer video. Gen-3 Alpha videos can be extended to a maximum of 40 seconds, where Turbo generations can be extended to a maximum of 34 seconds given that the original video was 10 seconds.
To extend a video, click the Use button and select Extend beneath an output in your session. Alternatively, you can extend an existing generation by opening it through your Assets and clicking Extend video.
The last frame of the video you’re extending will automatically populate as the input.
Add a new text prompt to indicate what should happen in the extension. Extensions are similar to Image to Video generations, so try keeping your prompt focused on camera, character and scene movement.
You can choose between a 5 or 10 second extension before generating in Gen-3 Alpha. The Turbo model offers 8 second extensions. Please note that extension costs will be the same as the pricing of the model used to generate the original video. The model cannot be changed before an extension.
Click Extend to begin the extension. If you’re happy with the extended output, you’ll be able to follow these steps up to two more times to create a long video.