Real Images

Tracking since Apr 1, 2026.

Changelog

• May 28, 2026
  description

  Real Images - Adobe Firefly Parallel Task Manager
  
  Real Images is a powerful Chrome extension that transforms your browser into a professional AI image generation workstation. It connects directly to Adobe Firefly's API infrastructure, enabling you to queue, manage, and execute bulk image generation tasks with multi-account support and parallel processing.
  
  Whether you are a content creator needing hundreds of product photos, a designer exploring visual concepts at scale, or a developer automating asset pipelines, Real Images gives you full control over the generation process without leaving your browser.
  
  
  FEATURES
  
  Multi-Model Support
  Generate images using multiple Adobe Firefly models from a single interface. Supported models include Gemini 3.1 (default), Gemini 2.5, Firefly Image 4, and Flux Ultra. Each model uses its own optimized API endpoint and payload structure.
  
  Parallel and Sequential Generation Modes
  Choose between sequential processing (one image at a time) or parallel processing (multiple images simultaneously across accounts). Turbo mode doubles the concurrency by running 2 jobs per account.
  
  Multi-Account Management
  Add, remove, and manage multiple Adobe IMS accounts. Each account stores its email profile, Bearer token, and API key. Accounts are persisted in chrome.storage.local and survive browser restarts.
  
  Automatic Adobe Token Capture
  Instead of manually copying Bearer tokens, the extension intercepts Authorization headers from your Adobe Firefly browser session using the webRequest API. Simply log into firefly.adobe.com, generate one image, then click Capture Token. The captured token is displayed in a list with copy and add-to-account buttons.
  
  20 Built-in Prompt Presets
  Choose from 20 professionally crafted prompt presets covering photography, 3D rendering, anime, UI/UX design, oil painting, watercolor, concept art, cyberpunk, fantasy, miniature worlds, and more. Presets are appended to your prompts automatically.
  
  Bulk Prompt Input
  Enter multiple prompts at once, one per line. Each line becomes a separate generation task. You can also load prompts from a .txt file using the Load TXT button.
  
  Google Search Grounding
  Toggle Google Search integration for Gemini models. When enabled, the model uses Google Search context to improve accuracy for prompts referencing real-world subjects, locations, or products.
  
  Flexible File Naming
  Control how generated images are named. Options include custom prefix, append mode (prefix + prompt), or replace mode (prefix only). Vietnamese diacritics can be preserved or stripped. All whitespace is normalized to hyphens. Filenames are capped at 150 characters to prevent filesystem errors.
  
  Reference Image Support
  Upload a reference image (displayed as a 60x60 thumbnail) to guide the generation process.
  
  Configurable API Endpoints
  Advanced users can override the default Generate and Upload API endpoints, as well as the Client API Key, directly from the Settings tab.
  
  Background Processing
  The extension uses a Manifest V3 Service Worker with a keep-alive mechanism that pings chrome.runtime every 15 seconds. This prevents Chrome from suspending the worker during long-running generation queues, even when the popup is closed.
  
  Dark Mode and Light Mode
  Toggle between dark and light themes with a single click. The preference is saved in localStorage and applied immediately on startup.
  
  Adjustable Font Size
  Use the font size slider (12px to 18px) to adjust the entire UI scale. The setting persists across sessions.
  
  Bilingual Interface (English and Vietnamese)
  The entire UI supports English and Vietnamese. Switch languages instantly with the language toggle button in the header. All labels, buttons, tooltips, preset names, and status messages update in real time. The language preference is saved in localStorage.
  
  Real-Time System Logs
  A dedicated Logs tab displays timestamped messages for every generation event including task start, completion, errors, and system status changes. The log buffer is capped at 200 entries to prevent memory issues.
  
  Progress Tracking
  A progress bar and status text show the current state of the generation queue (e.g., Completed 5/20). The idle/running badge in the header provides at-a-glance status.
  
  Smart Download Pipeline
  Generated images are downloaded directly to a configurable subfolder inside your Downloads directory using the chrome.downloads API. Each file receives a unique timestamp and random suffix to prevent collisions.
  
  Async Job Polling
  After submitting a generation request, the extension polls the job result URL with exponential backoff (starting at 800ms, max 3000ms, up to 40 attempts) until the image is ready.
  
  Round-Robin Task Distribution
  In parallel mode, tasks are distributed evenly across active accounts using round-robin allocation. In turbo mode, each account is duplicated as a virtual worker to double throughput.
  
  
  HOW IT WORKS
  
  Step 1: Add your Adobe account. Go to the Settings tab, click Open Firefly to log into your Adobe account in a new tab. Generate one image on the Firefly website. Return to the extension and click Capture Token. The extension automatically intercepts the Bearer token from your browser session. Click the add button on the captured token to save it as an account.
  
  Step 2: Configure your generation settings. On the Image tab, select your preferred model, aspect ratio, resolution, and optionally enable Google Search grounding. Choose a prompt preset if desired.
  
  Step 3: Enter your prompts. Type or paste your prompts into the text area, one per line. Each line becomes a separate generation task. You can also load prompts from a .txt file.
  
  Step 4: Start generation. Click Start Gen. The extension distributes tasks across your accounts and begins generating images in the background. You can close the popup and the generation continues in the Service Worker.
  
  Step 5: Monitor progress. Open the Logs tab to see real-time status updates. The progress indicator in the header shows completion count.
  
  Step 6: Collect your images. Generated images are automatically downloaded to the folder you specified (default: Downloads/RealImages).
  
  
  FREQUENTLY ASKED QUESTIONS
  
  Q: Do I need an Adobe Creative Cloud subscription?
  A: You need an Adobe account with access to Adobe Firefly. The extension uses your existing session tokens to authenticate API requests.
  
  Q: Can I run generation with the popup closed?
  A: Yes. The Service Worker continues processing in the background. The keep-alive mechanism prevents Chrome from suspending it.
  
  Q: How many images can I generate at once?
  A: There is no hard limit in the extension. You can queue as many prompts as you want. The practical limit depends on your Adobe account quotas.
  
  Q: Is my token stored securely?
  A: Tokens are stored in chrome.storage.local, which is sandboxed to the extension and not accessible by websites or other extensions.
  
  Q: Can I use multiple Adobe accounts?
  A: Yes. Add as many accounts as you need. In parallel mode, tasks are distributed across all active accounts automatically.
  
  
  SUPPORT
  
  For bug reports, feature requests, or questions, contact: [email protected]

  Real Images — Professional AI image generation in Chrome
  
  Real Images turns your browser into an AI image orchestration workstation: you enter ideas (prompts), configure marketing/composer fields, and choose output modes — the extension sends requests to your already logged-in AI provider tabs  and saves generated images to disk via your Downloads folder.
  
  Core difference: The extension does not ask you to paste API keys or session tokens into its settings. Authentication relies on your normal logged-in sessions in each provider tab (cookies/session). The extension only automates the UI and captures image/text results within those tabs.
  
  Key features
  
  AI-to-AI flow: Optimize prompts with one AI (e.g. Claude/ChatGPT) and generate images with another (e.g. Gemini/Grok), or use a single AI for both — configurable.
  Multiple providers: Works with the AI platforms supported in the extension catalog, with separate picks for prompt AI, image AI, and optional metadata AI.
  Bulk runs: Many prompt lines; concurrency is managed per provider to reduce rate limits and duplicate outputs.
  Output modes: Banner, icon, logo, product, infographic — with layout/text directives suited to each mode.
  Reference images & watermark: URL or data: references; logo watermark with position/opacity; optional logo guidance for generation where applicable.
  Visual Style Lock: Lock palette, lighting, mood, typography, etc. for consistent batches.
  Brand Profile: Brand fields can be injected into AI prompts (optional) and used for metadata / embedded image metadata.
  Metadata & files: Optional SEO-oriented metadata; downloads use safe filename rules (Unicode-aware, illegal filesystem characters stripped).
  UI: Side panel workflow (no popup), dark/light theme, English/Vietnamese, adjustable font size.
  In-panel logs: Batch progress feedback (production builds may tone down verbose logging per build flags).
  How to use (quick)
  
  Install the extension and open the Real Images side panel.
  Log in to the AI sites you plan to use and pin the tabs you want as stable targets (or follow the extension’s tab guidance).
  In the panel: choose flow (direct / split / simple), assign prompt AI and image AI, enter prompts (single or multi-line).
  Optionally enable watermark, reference image, Visual Style Lock, Brand Profile, metadata.
  Start generation — the extension drives the relevant tab, waits for the response, and saves images locally.
  Permissions & privacy
  
  The extension uses storage for settings and flow state; downloads to save images; tabs, scripting, sidePanel for orchestration and the side panel UI; host_permissions are scoped to AI provider domains declared in the manifest (not expanded arbitrarily).
  The extension is not designed to collect or ship your prompts, images, or login tokens to publisher servers; processing happens on your device and in your open AI tabs. You must still comply with each provider’s terms and acceptable-use policies.
  Notes
  
  Quality and speed depend on your plan/account and each service’s image generation capabilities; some AIs cannot generate images — the extension reflects supported capabilities in its UI.
  Provider UI changes over time can affect automation; updates may be required.
  Support
  
  Use the developer contact email shown on the store listing / manifest for bug reports or technical questions.

• May 28, 2026
  short_description

  Premium prompt engineering and Adobe Firefly API parallel task manager.

  AI-to-AI image flow: generate prompts with one AI, generate images with another (or use a single AI for direct image generation).

• May 28, 2026
  host_permissions

  *://*.adobe.io/*, *://*.adobe.com/*, *://ims-na1.adobelogin.com/*

  *://chatgpt.com/*, *://chat.openai.com/*, *://claude.ai/*, *://gemini.google.com/*, *://grok.com/*, *://www.perplexity.ai/*, *://perplexity.ai/*, *://firefly.adobe.com/*, *://ims-na1.adobelogin.com/*, *://*.oaiusercontent.com/*, *://*.googleusercontent.com/*, *://*.adobe.com/*, *://*.adobe.io/*, *://*.perplexity.ai/*, *://imagedelivery.net/*, *://*.grok.com/*

• May 28, 2026
  permissions

  storage, downloads, unlimitedStorage, webRequest

  storage, downloads, unlimitedStorage, sidePanel, scripting, tabs

Permissions & access

Permissions

storagedownloadsunlimitedStoragesidePanelscriptingtabs

Host access

*://chatgpt.com/*, *://chat.openai.com/*, *://claude.ai/*, *://gemini.google.com/*, *://grok.com/*, *://www.perplexity.ai/*, *://perplexity.ai/*, *://firefly.adobe.com/*, *://ims-na1.adobelogin.com/*, *://*.oaiusercontent.com/*, *://*.googleusercontent.com/*, *://*.adobe.com/*, *://*.adobe.io/*, *://*.perplexity.ai/*, *://imagedelivery.net/*, *://*.grok.com/*

Screenshots