Feed it images, web pages, and your voice

🎯Goal. Use Aider's richer inputs — images, web pages, and voice — to give the model better context with less typing.

1. 1Aider accepts images and web pages as context. Hand it a screenshot of a chart or a doc page and ask it to match or implement what's shown — the model sees the reference, not just your description.
2. 2Use voice-to-code: dictate your request instead of typing it. Handy for longer instructions or when your hands are busy at the keyboard.
3. 3Remember it supports 100+ programming languages — Python, JavaScript, Rust, Go, C++, PHP, HTML, CSS and more — so the same workflow carries across whatever stack your project uses.

✓You'll see. Aider acting on an image or web page you provide, or on a spoken instruction — turning richer context into committed code changes.

💳Cost. Still just LLM token cost; image and web-page context add to the tokens per request, so keep references focused.

💡Takeaway. Aider isn't limited to typed prompts in one language — images, web pages, voice, and 100+ languages all feed the same Git-committed workflow.