Adobe expands Firefly with AI video editing tools and partner model integrations

By NCS Staff December 16, 2025

Weekly insights on the technology, production and business decisions shaping media and broadcast. No paywall. Independent coverage. Unsubscribe anytime.

Adobe has introduced new capabilities to its Firefly platform, including text-based video editing, camera motion controls and expanded access to third-party AI models. The updates are designed to give users greater precision and flexibility in generative video workflows.

A new “Prompt to Edit” feature allows users to apply specific natural language instructions to existing AI-generated video clips, using Runway’s Aleph model.

Rather than regenerating clips from scratch, users can refine them with targeted changes such as removing objects, altering backgrounds, or zooming in on a subject. Edits are applied directly to the original clip, preserving previous creative choices.

Firefly also now supports camera motion reference uploads. Users can upload a start frame and a reference video, enabling the Firefly Video Model to recreate similar camera movements in the generated scene.

Adobe has expanded its model offerings within Firefly to include Topaz Labs’ Astra model for video resolution upscaling. The tool allows users to enhance video quality to 1080p or 4K, including restoration of low-resolution or archival footage. Upscaling is integrated into Firefly Boards, allowing batch processing while maintaining active workflows.

The platform has also integrated FLUX.2, a new image model from Black Forest Labs. FLUX.2 supports photorealistic rendering, flexible aspect ratios, advanced text generation, and reference-based image synthesis. It is now available in Firefly’s Text to Image module, Prompt to Edit, Firefly Boards, and Photoshop desktop. Availability in Adobe Express is scheduled for January.

Adobe has launched the public beta of Firefly video editor, a browser-based editing tool that combines AI-generated media with user content. The editor includes a timeline for detailed edits and a text-based editing mode for rearranging dialogue or interviews through transcript editing. Final videos can be exported in multiple formats.

To support exploration of the new features, Adobe is offering unlimited image and video generations through Jan. 15 for customers on Firefly Pro, Firefly Premium, and select credit-based plans.

Advertisement

This includes use of Firefly’s native models and partner models such as FLUX.2, Google’s Nano Banana, and OpenAI’s GPT Image.

The Firefly platform continues to serve as Adobe’s unified space for generative content creation across video, audio and imagery.