Adobe Firefly Adds Luma AI Ray 3.14; Supports “Modify Video” for AI Video Editing

On Friday, February 6, Adobe added Ray 3.14 — the latest video generation model from Luma AI — to the Firefly partner model lineup.

Adobe Firefly has added Ray 3.14, Luma AI’s latest cutting-edge model released in January 2026.

To celebrate the launch, a limited-time unlimited generation campaign is currently underway.

New Partner Models & Credit Costs:

Video Generation Models:

Ray 3.14 - Luma AI : 40 credits/sec (720p)

Ray 3.14 HDR - Luma AI : 120 credits/sec (720p)

Ray 3.14 offers significant improvements in generation speed and cost-efficiency compared to its predecessor, Ray 3.

Adobe Firefly now supports video generation at up to 4K resolution with a maximum duration of 10 seconds with Ray 3.14.

Additionally, users can now access Modify Video, the video-to-video (v2v) editing feature of the Ray series.

To access this, simply select "Prompt to edit clip" from the top-left menu of a generated video to enter the editing interface.

Ray 3.14 is available via Firefly Generate Video, Firefly Boards, and the Firefly mobile app.

Note: A Firefly plan or Creative Cloud Pro subscription is required to use partner models.

A limited-time campaign for unlimited video generation with Ray 3.14 is running until Wednesday, February 18 (PT).

Eligible Plans for Unlimited Access:

・Firefly Standard, Pro, Premium

・2,000, 4,000, 7,000, 50,000 credits (Credit Add-on Plans)

Free users: Up to 5 free generations available.

Exclusions:

* Creative Cloud Pro users are not eligible for the unlimited offer.

* The Ray 3.14 HDR model is not included in the unlimited campaign.

For those looking for the best value, upgrading to Firefly Pro (4,000 credits) or higher is recommended, as these plans also include unlimited image generation.

It's a great time to jump in and experience these new capabilities.

関連記事 : Adobe Firefly Image 5 画像生成AIモデルをリリース 画像編集にも対応

Adobe Firefly offers a comprehensive suite of features on its web browser version.

It provides access to a wide range of generative AI capabilities—including image, vector, and video generation—and allows users to utilize various partner models.