Adobe Firefly
Adobe's AI studio for image, video and audio generation
Editorial take
Adobe Firefly fits best if you already work in Adobe's apps. One plan gives you Adobe's own image and video models, which Adobe trains on licensed and public domain content, alongside partner models from Google, OpenAI, Runway, Kling and others, and the same credits work in Photoshop, Premiere and Express. Standard image generation is unlimited on paid plans, but video, audio and partner models draw on monthly credits, so heavy video users should compare credit allowances before picking a tier. The newest pieces announced on November 10, 2026, Firefly Builder and the local MCP integrations, are still in private beta or preview.
What is Adobe Firefly?
Adobe Firefly is Adobe's generative AI app for images, video, audio and vector graphics, available on the web at firefly.adobe.com and as a mobile app for iOS and Android. It generates images from text and edits them with prompts (Generative Fill, Generative Expand, background removal and edits to objects, text and lighting), creates vector graphics, generates video from text or images with camera controls, and handles audio work such as sound effects, music, speech and translating audio or video into other languages. Firefly Boards is a shared canvas for mood boards and storyboards, the Firefly video editor assembles clips on a timeline with captions, Firefly Graph (beta) builds node-based workflows that chain models and Adobe tools, and Firefly AI Assistant (beta), powered by the Adobe Creative Agent, takes requests in plain language.
Firefly mixes Adobe's own models, such as Firefly Image Model 5 and the Firefly Video Model, with partner models. Adobe's help pages list image and video options from Google (Nano Banana, Imagen and Veo), OpenAI (GPT Image), Black Forest Labs (FLUX), Runway, Luma AI, Kling, ByteDance (Seedance), Ideogram and Topaz Labs, and Adobe also names ElevenLabs as a partner. Adobe trains its own Firefly models on licensed and public domain content, says their output is safe for commercial use and attaches Content Credentials, while users are responsible for checking whether partner model output suits their project. Generative credits from a Firefly plan also work in Photoshop, Illustrator, Premiere, Lightroom and Adobe Express, and Adobe connectors bring Photoshop, Lightroom, Express and Firefly tools into Google Gemini, Claude and ChatGPT.
At Adobe MAX on November 10, 2026, Adobe announced Firefly Builder, a private beta with a waitlist that turns a plain-language description into an interactive prototype of a website, presentation or app; TechCrunch framed it as a move into Figma's territory. Adobe also released Firefly Design (beta), a canvas for illustration, image editing, animation, 3D and print, made Edit Text, Relight, Retouch and Quick Actions generally available in Firefly, and previewed local MCP integrations that will let third-party AI agents edit native Photoshop, Illustrator, Premiere and After Effects files, with a private beta planned for early 2027. The same day, Adobe and Google expanded their partnership: Google's latest image and video models are coming to Firefly, Photoshop, Premiere and Express (Gemini Omni, Veo and Nano Banana are already available), Adobe's Gemini connectors will add Acrobat and more creative tools, and Acrobat will appear as a suggested app in Google Drive and Gmail.
The free plan gives a limited number of daily generations on a curated set of models, plus the image and video editors and Firefly Boards. Paid plans include unlimited standard image and vector features such as Generate Image and Generative Fill, and use monthly generative credits for video, audio and partner models. Firefly Standard costs $9.99 per month with 2,000 credits, Firefly Pro $19.99 with 4,000 credits plus Photoshop on web and mobile and Adobe Express Premium, Firefly Pro Plus $49.99 with 10,000 credits, and Firefly Premium $199.99 with 50,000 credits and unlimited use of the Firefly Video Model in Generate Video. Credits reset each billing cycle and do not roll over, and Pro, Pro Plus and Premium are also sold per license for teams. Adobe is also running a special offer, which it says may change, that gives new individual Pro, Pro Plus and Premium subscribers unlimited generations on select models for their first year.
Best for
Designers, marketers and creators already working in Photoshop, Express or Creative Cloud who want image, video and audio generation in one plan
Key strength
Adobe's own models, trained on licensed content, alongside partner models from Google, OpenAI and others in one app
Score breakdown (out of 5)
What you would use it for
- →Generating and editing marketing images and social graphics
- →Producing short videos, b-roll and storyboards from text or images
- →Comparing Adobe, Google and OpenAI image models in one workspace
- →Translating video and audio into other languages for localized content
- →Extending or retouching photos before finishing them in Photoshop
Pros & Cons
👍 Pros
- ✓One subscription covers Adobe's own models plus partner models from Google, OpenAI, Runway, Kling and others
- ✓Adobe trains its Firefly models on licensed and public domain content and positions their output as safe for commercial use
- ✓Generative credits also work in Photoshop, Illustrator, Premiere, Lightroom and Express
- ✓Free plan with daily generations, and unlimited standard image features on every paid plan
👎 Cons
- ✗Video, audio and partner models draw on monthly credits that do not roll over
- ✗Photoshop and Express Premium access starts at the $19.99 Pro plan
- ✗Adobe's commercial-safety position does not extend to partner model output
- ✗Firefly Builder and the local MCP integrations are not generally available yet
Key Features
- ✓ Text to image with Firefly Image Model 5 and partner models
- ✓ Generative Fill, Generative Expand and prompt-based edits to objects, text and lighting
- ✓ Text and image to video with the Firefly Video Model and partner video models
- ✓ Text to vector graphics
- ✓ Sound effects, music and speech generation, plus audio and video translation
- ✓ Firefly Boards for mood boards and storyboards
- ✓ Firefly video editor with a timeline, captions and generated b-roll
- ✓ Firefly AI Assistant (beta) and Firefly Graph (beta) node workflows
- ✓ Firefly Design (beta) canvas and Firefly Builder prototypes (private beta)
- ✓ Content Credentials on generated content, with custom models and IP indemnification on enterprise plans
Available on
Integrates with
Adobe Firefly Pricing
✅ Adobe Firefly has a free plan - no credit card required to start.
Free
- ✓Limited free daily generations for images, video and audio
- ✓Limited access to top models, including Nano Banana and Luma Ray 3.14
- ✓Image and video editors
- ✓Firefly Boards
Firefly Standard
- ✓2,000 generative credits per month
- ✓Unlimited standard image features like Generative Fill
- ✓Up to 20 five-second videos
- ✓Translate up to 6 minutes of audio and video
- ✓Up to 200 sound effects and 100 minutes of music
- ✓Partner models from Google, OpenAI and more
- ✓Adobe Fonts
Firefly Pro
- ✓4,000 generative credits per month
- ✓Up to 40 five-second videos
- ✓Translate up to 13 minutes of audio and video
- ✓Up to 400 sound effects and 200 minutes of music
- ✓Adobe Express Premium
- ✓Photoshop on web and mobile (full access)
Firefly Pro Plus
- ✓10,000 generative credits per month
- ✓Up to 100 five-second videos
- ✓Translate up to 33 minutes of audio and video
- ✓Up to 1,000 sound effects and 500 minutes of music
- ✓Firefly Graph (beta)
- ✓Adobe Express Premium and Photoshop on web and mobile
Firefly Premium
- ✓50,000 generative credits per month
- ✓Unlimited access to the Firefly Video Model in Generate Video
- ✓Translate up to 166 minutes of audio and video
- ✓Up to 5,000 sound effects and 2,500 minutes of music
- ✓Adobe Express Premium and Photoshop on web and mobile
From the blog
Developer resources
Related Tools
Midjourney consistently produces the most aesthetically refined AI images of any tool in this category. Its outputs have a distinctive visual quality that distinguishes them from DALL-E or Stable Diffusion. The web app is now the main interface and Discord remains available. The prompting style takes some practice, but the community gallery and fast iteration loop help new users learn it.
ChatGPT is the most capable all-in-one AI assistant available. Its combination of image generation, voice mode, code execution, web browsing, and a large custom GPT ecosystem makes it the default choice for users who want one tool that handles everything. The tradeoff is that it is rarely the deepest tool on any single task.
Runway leads the AI video generation space for creative professionals. Gen-4.5 produces cinematic clips from text or images, and Aleph 2.0 in Edit Studio lets you edit one frame of existing footage and carry the change through the clip, while Act-Two handles performance capture and apps cover object removal and background swaps. Third-party models such as Kling and Seedance fill gaps like native audio. It is expensive for casual use but the right tool for serious video work.
DALL-E 3 was retired in 2026 and replaced by OpenAI's GPT Image models, now on the Images 2.6 release with Sunburst and Flare tiers. Image generation inside ChatGPT remains the lowest-friction path to AI images for existing ChatGPT users. For professional-grade aesthetics, Midjourney still leads.
This page contains affiliate links. We may earn a commission at no extra cost to you. Learn more.
