Partner models to generate videos

Last updated on Sep 4, 2025

Learn how to generate videos in Firefly Boards using partner models.

While working in Firefly Boards (beta), you can generate videos and place them directly onto the canvas. You can choose from Adobe models like Firefly Video or partner models such as Veo 2, Veo 3, Pika 2.2, and Ray2, and Runway Gen-4 Video.

Note

The availability of video generation settings depends on the video model selected.

Generate a video using Veo 2 and Veo 3

On the Firefly homepage, select Boards (beta).

On the Boards page, select Create new board.

Select Describe what you want to generate at the bottom of the page.

Select the Video tab to view all the generation settings in the prompt bar.

Select the video model as Veo 2 or Veo 3.

The Video option is selected in the prompt bar and from the list of Adobe and Non-Adobe models, Veo 3 is selected. — Choose the Veo 3 model for video generation with complementary audio.

If you've selected the Veo 2 model, you can add an image under the First frame option to use as the starting frame of the generated video.

If images have already been added to the canvas and you want to use one as the first frame, select the Sample from canvas option, then use the eyedropper tool to choose the image. From the context menu, select Use as image reference, and optionally add a sub-prompt to the main prompt.

The image added to the canvas is chosen after the Sample from canvas is selected. The context menu is open and the Use as image reference is selected. — Select the Use as image reference option to use an image from the canvas as the first frame for video generation.

Alternatively, you can also use the Upload image reference option to upload a reference image.

Select the aspect ratio for your generated video—Landscape, Portrait, Square, or Widescreen.

Enable the Audio option to generate synchronized sounds or music for the generated video.

The Veo 3 model is selected for video generation and the Audio option is enabled in the prompt bar. — Use the Audio option to generate audio along with the video.

Note

Veo 3 is the only model that offers the ability to generate audio along with your video.

To use additional generation settings, select the Generation settings icon next to the Aspect ratio options. This activates the right-side panel, which includes the following additional settings:

Sub-prompts: You have to option to browse through a range of sub-prompts from lighting to styles. When you select the preview thumbnails of the sub-prompts, they get added to the prompt bar.

Seed: You can add a seed number that helps start the process and controls the randomness of what the AI creates. If you use the same seed, prompt, and control settings, you can generate similar video clips.

Add a text description in the prompt bar.

Select Generate.

The generated video is selected and there is an option to add it to the canvas. — Preview the generated video and use the drag-and-drop action to add it to the canvas.

Generate a video using Pika 2.2 and Ray2

Before you begin:

Create a new board, then select Describe what you want to generate at the bottom of the page to start generating a video.

Select the Video tab to view all the generation settings in the prompt bar.

Select the video model as Pika 2.2 or Ray2.

If you have images that you want to use as the first and last frames of the generated video, add an image under the First frame and Last frame, respectively.

Pika 2.2 is selected and in the prompt bar, there are options to add first and the last frame for video generation. — Add images as the first and last frames, or select them from the canvas for video generation.

These uploaded images for the first and last keyframes serve as fixed points, providing guidance to ensure the video starts and ends with them.

If images have already been added to the canvas and you want to use one as the first frame, select the Sample from canvas option, then use the eyedropper tool to choose the image.

The image used for first frame is visible in the canvas. One of the images is selected and from the context menu, the Use as last frame option is selected. — Select an existing image in the canvas to add it as the first or last frame of the generated video.

Alternatively, you can also use the Upload image reference option to upload a reference image.

Select the aspect ratio for your generated video—Landscape, Portrait, Square, or Widescreen.

To use additional generation settings, select the Generation settings icon next to the Aspect ratio options.

Add a text description in the prompt bar.

Select Generate.

The generated video is highlighted and it is ready to be added to the canvas. — Use the drag-and-drop action to add the generated video to the canvas.

Generate a video using Runway Gen-4 Video

Before you begin:

Create a new board, generate or add an image to the canvas, then select Describe what you want to generate at the bottom of the page to start generating a video.

Select the Video tab to view all the generation settings in the prompt bar.

Select the video model as Runway Gen-4 Video.

Select the Sample from canvas or Image (required) option, then select the image from the canvas with the eyedropper tool. From the context menu, select Use as image reference, and optionally add a sub-prompt to the main prompt.