Higgsfield Reframe Changes Video Aspect Ratios by Regenerating Backgrounds Instead of Cropping

HiggsfieldHiggsfield

· Updated

Higgsfield AI launched Reframe, a tool that uses generative AI to adapt videos to different aspect ratios by outpainting the background rather than cropping the frame. By integrating with the Model Context Protocol, the tool allows Claude and other AI agents to autonomously repurpose widescreen footage into vertical formats while keeping all subjects in view.

Higgsfield AI, a platform for hosting frontier video models, released Reframe on its web platform and via its Higgsfield's Model Context Protocol server. The tool uses generative AI to change aspect ratios by outpainting instead of traditional cropping.
Max clip length
1 minute
Supported ratios
9:16, 16:9, 1:1, and others
Reference images
Up to 4
Starter pricing
$15 per month (billed annually)
Access methods
Web, Claude MCP, Adobe plugins

Traditional reframing tools often cut off secondary subjects or essential background details when converting widescreen content for social media. This update follows the launch of Higgsfield's Adobe Premiere plugins but moves the capability into the agentic loop. It enables AI assistants to handle complex visual transformations that previously required manual editing.

You can upload clips up to one minute long and choose from ratios like 9:16, 21:9, or 1:1. The system supports up to four reference images to maintain visual consistency in the regenerated areas. Reframe is available now on the web and for Claude via MCP, with several subscription tiers available.

View the full update on higgsfield.ai
Higgsfield AI 🧩
Higgsfield AI 🧩
@higgsfield
X

Сlaude can now change aspect ratios without cropping out half your shot. Meet Higgsfield Reframe. Give it a reference image and the background holds up in any format. Works for podcasts, IRL streams, commercials, films, and cartoons. On Higgs, or in Claude via Higgs MCP. https://t.co/RQ5iNtDluC

104retweets831likes
View on X

Still wondering? A few quick answers below.

Higgsfield Reframe is an AI-powered video tool that adapts footage to different aspect ratios without using traditional cropping. Instead of cutting off the edges of a shot, the system uses generative AI to outpaint and regenerate the missing background areas. This ensures that all subjects remain in the frame when converting between widescreen and vertical formats.

Most reframing tools simply crop the edges of a video, which often results in losing people or details on the sides of the shot. Higgsfield Reframe analyzes the scene to lock onto subjects and then regenerates the surrounding environment to fit the new ratio. This process maintains the integrity of multi-person scenes without stretching or blurring the footage.

The tool supports a wide variety of formats including 9:16, 16:9, 4:3, 3:4, 21:9, and 1:1. These options cover the standard requirements for social media platforms like TikTok and Instagram Reels, as well as cinematic widescreen and square layouts. Users can upload any video up to one minute in length to be reframed into these formats.

You can use Reframe directly through the Higgsfield web platform or as a native plugin for Adobe Premiere Pro and After Effects. Additionally, it is available as a tool for Claude and other AI agents via the Higgsfield Model Context Protocol server. This integration allows users to trigger video reframing tasks using natural language instructions within an AI chat.

Reference images are optional files that users can upload to guide the AI during the regeneration process. By providing up to four images, you can ensure that the newly generated background areas match the specific look, lighting, and style of your original footage. This helps maintain visual consistency across the extended canvas of the reframed video.

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

Higgsfield

See all AI news & updates from Higgsfield →

Share this update