Anki Gemini Vision is a powerful Anki add-on that integrates Google’s Gemini 2.5 Flash model directly into your Anki editor. It allows you to right-click any image in your card editor to seamlessly extract text, complex math formulas (KaTeX format), and structured lists, converting them instantly into editable Anki text!
✨ Features / 功能特点 🤖 AI Image to Text: Right-click an uploaded image and select [Gemini] 将此图片内容转为文本附加到末尾 to extract its text. 🧮 KaTeX & Math Formula Support: The prompt is specifically optimized to output rendering-friendly Anki KaTeX formats ($…$ for inline, $$…$$ for block display without breaking HTML tags). 📋 Smart Formatting: Recognizes and structures complex lists, options, and multi-paragraph text without relying on unsupported Markdown tables in Anki. ⚡ Local & Private: Uses your own Google Gemini API Key. Fast and secure. 🌐 Proxy Support: Built-in HTTP/SOCKS5 proxy configuration for regions where Gemini API is restricted. 🚀 How to Use / 使用方法 Get an API Key: Go to Google AI Studio to generate a free Gemini API Key. Configure Add-on: In Anki, go to Tools -> Add-ons -> Select Anki Gemini Vision -> Config. Fill in your “api_key”. If you need a proxy, fill in “proxy_url” (e.g., http://127.0.0.1:7897). Use in Editor: Open any card in the Add or Browser window. Paste or Insert an image. Right-click on the image. Click the Gemini extraction button. Wait a few seconds, and the recognized text will be appended automatically! 🛠️ Configuration Example / 配置文件示例 { “api_key”: “YOUR_GEMINI_API_KEY_HERE”, “proxy_url”: “http://127.0.0.1:7897” }
🐛 Troubleshooting / 常见问题排查 HTTP Error 400 (User location is not supported): Your current IP is in a region not supported by Google Gemini. Please enable the proxy software, ensure it’s running, and fill in the proxy port in the add-on configuration. No Response / Plugin error: Double-check that your API key is valid and has sufficient quota.
Developed using Antigravity AI Assistant. Open Source / Feedback welcome!
