Artemis Vision

I wanted to join Google's #withlovegemini challenge — generate an image with Gemini and share it on social media. Simple, right?

Not quite. I'm blind. I use a screen reader to navigate everything on my computer. When Gemini generated my image, LinkedIn rejected it: wrong dimensions. The obvious fix is to crop it, but the crop tools on LinkedIn and Instagram are completely inaccessible with a screen reader. No labels, no keyboard support, nothing.

So I built this tool. It calls the Gemini API directly and generates images already sized for each platform. No cropping needed. Problem solved.

You don't need sight to have vision — and technology is only powerful when it works for everyone.

Want to know the full story? Read more on GitHub

How does it work?

Step 1: Get your API key

Go to Google AI Studio and create a free API key. It takes less than a minute.

Step 2: Set your key

Click the "Set API Key" button at the top of this page and paste your key. It's stored only in your browser session and never sent anywhere except to the Gemini API.

Step 3: Describe your image

Write a prompt describing what you want to generate. Be as specific as you like — style, colors, composition, mood. You can also upload reference images for guidance.

Step 4: Choose your settings

Pick the model (Pro for quality, Flash for speed), the aspect ratio for your target platform (9:16 for Stories, 16:9 for LinkedIn, etc.), and the resolution.

Step 5: Generate

Click "Generate Image" and wait. The app will create your image and automatically generate an accessible description (alt text) for it. You can then download the image and copy the alt text to share it on social media.

Generate an Image

Upload up to 14 reference images. Accepted: PNG, JPEG, WebP, GIF.

    Describe what you want the AI to create. Be specific about style, content, and composition.

    Choose the aspect ratio. 9:16 for Instagram Stories and Reels. 4:5 for Instagram feed. 16:9 for LinkedIn.