fal Launches ByteDance Bernini-R for Unified Video Generation and Editing

falfal

fal has launched ByteDance's Bernini-R model on its platform. This unified model combines video generation and instruction-based editing, allowing users to modify elements like objects or styles while maintaining scene consistency, and guide outputs with up to five reference images.

fal has made ByteDance's Bernini-R model available, a unified system for both generating and editing video content. This model allows users to create video from text prompts and perform instruction-based edits, such as swapping objects, changing weather, or adjusting backgrounds, while preserving the original scene.
Model Owner
ByteDance
Pricing
$0.08/second (848px video)
Reference Image Support
Up to 5 images
Core Capabilities
Video generation, video editing
Resolution Scaling
0.5x for ≤576px, 1x for 848px, 2x for 1280px

Integrating video generation and editing into a single model streamlines the creative process, reducing the need for multiple specialized tools. The ability to guide video output with up to five reference images also helps maintain visual consistency across generated or edited clips.

Bernini-R is available on the fal platform for commercial use, with pricing at $0.08 per second for 848px video, scaling by resolution. You can use it for text-to-video generation, instruction-based video editing, and reference-guided video creation.

View the full update on fal.ai
fal
fal
@fal
X

Bernini-R is now live on fal! One unified model that both generates and edits video Edit by instruction: swap objects, weather, backgrounds, camera angles or style, scene stays intact Guide it with up to 5 reference images for a consistent look https://t.co/hDRvKHee7U

14retweets141likes
View on X

Still wondering? A few quick answers below.

Bernini-R is a unified AI model developed by ByteDance that can both generate and edit video content. It is designed to create high-quality video from text prompts and modify existing videos based on instructions.

Bernini-R supports instruction-based editing, allowing you to change elements like objects, weather, backgrounds, camera angles, or overall style within a video. The model is designed to keep the original scene intact during these modifications.

The model can be guided by up to five reference images. These images help maintain a consistent look and style throughout the generated or edited video, ensuring visual coherence across different segments.

Bernini-R is available on the fal platform. You can use it via API for commercial applications or through the web playground for direct interaction.

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

Share this update