Kavel AI Image Generator for Obsidian

Put a picture in your note without leaving it. Run Generate image into this note, describe what you want, and the image is saved to your vault and embedded at the cursor. Right-click any image in your vault and choose Edit with Kavel to change it with one sentence.

No API key needed to start — powered by Kavel.

What you can do

• Generate — prompt + aspect ratio (1:1, 16:9, 9:16, 4:3, 3:4). The image lands in the Kavel folder (configurable) and ![[…]] is inserted where you were typing.
• Edit — from the file menu of any PNG, JPG or WebP, or with the cursor on an embedded image: "make the sky a warm sunset; keep everything else unchanged". The edited copy is saved next to the others.
• Works on desktop and mobile.

Free tier, then your account

With the API key field empty the plugin runs on Kavel's free tier: no account, 1K output with a watermark, a few images a day. Paste a key from kavel.ai/settings/apikeys into Settings → Kavel and every run uses your account instead — your credits, no daily ceiling, no watermark on a paid plan, and the model of your choice: Nano Banana 2, GPT Image 2, Qwen Image 3 for pictures with readable text, Seedream 5.0 Pro. Pricing lists the plans.

Prompts that work

Name the style, the light and what is in frame: "watercolor map of a small coastal town, soft morning light" beats "map". For edits, say what must stay the same — the edit model (Nano Banana 2 Lite) keeps what it can see and invents what it cannot.

Privacy and network use

This plugin sends your prompt to kavel.ai to generate images. When you edit an image, that image is uploaded to Kavel's CDN so the model can read it. Nothing else in your vault is read or sent. No telemetry.

MIT