Adobe Firefly | New AI video tools add editing and upscaling
Adobe has expanded Firefly with new AI video creation tools, new partner models, and a limited-time unlimited generation offer for eligible plans. Published on December 16, 2025, the update adds Prompt to Edit controls for video, camera motion reference, Topaz Astra upscaling in Firefly Boards, FLUX.2 support, and the broad public beta of Firefly video editor.
Adobe Firefly gives creators more control over AI video production
Adobe’s update focuses on a key problem in generative video: creators often get a clip that is close to useful, but not precise enough for production. Instead of forcing users to regenerate the entire result, Firefly is adding tools that let creators refine a generated clip, control camera movement, improve resolution, and continue editing in a dedicated timeline.
For designers, editors, and visual production teams, this matters because AI video workflows need more than generation. Professional use also depends on correction, upscaling, sequencing, audio, export control, and the ability to preserve a useful result while adjusting the details that do not work.
How Firefly improves AI video editing
The new Prompt to Edit controls for video let users refine an existing Firefly-generated clip with text instructions. Adobe says the feature uses Runway’s Aleph model to make targeted changes, such as removing an object, replacing a background, changing the sky, lowering contrast, or zooming in slightly on the main subject.
Firefly is also adding camera motion control. Users can upload a starting frame and a reference video that shows the camera movement they want to recreate. This gives creators more control over cinematic movement while keeping the generated scene anchored to the intended visual direction.
New production options for AI video workflows
Topaz Astra is now available in Firefly Boards for video upscaling, allowing creators to push footage to 1080p or 4K. That can help with low-resolution clips, older footage, archival material, or generated assets that need to be sharp enough for social platforms, presentations, YouTube formats, or client review.
The Firefly video editor is also now broadly available in public beta. Adobe describes it as a browser-based creative assembly space where users can combine generated clips, music tracks, visuals, and their own footage in a lightweight multi-track timeline, with options to edit by timeline or by transcript text.
Adobe also adds FLUX.2 from Black Forest Labs to Firefly, with access in Text to Image, Prompt to Edit, Firefly Boards, Photoshop desktop’s Generative Fill, and Adobe Express starting in January. For visual teams, this expands model choice across image generation, editing, references, and AI-assisted design exploration.
Availability and unlimited generations
Adobe says the limited-time unlimited generation offer runs until January 15 for customers on Firefly Pro, Firefly Premium, 7,000-credit, and 50,000-credit plans. The offer includes unlimited image and video generations in the Adobe Firefly app for eligible users.
The promotion covers Firefly’s video model, Firefly image models, and partner image models, including Black Forest Labs’ FLUX.2, Google’s Nano Banana, and OpenAI’s GPT Image. For production use, designers should still review quality, licensing, resolution, continuity, export settings, and client requirements before using AI-generated video assets in final work.
Sources and Recommended Links
- Adobe Firefly improves AI video creation with new tools, new models and unlimited generations | Adobe Blog (Official)
- Adobe Firefly | Adobe (Official)
- Firefly video editor | Adobe (Official)
