Delphic

Local AI that runs entirely in your browser.
No server. No API key. No data leaves your device.

Qwen3-0.6B · ~400 MB · one-time download

Cached locally — instant on repeat visits

Fully offline after first load

~400 MB on first use, then cached forever via OPFS

Qwen3-0.6B

Delphic Ready · Qwen3-0.6B · on-device

Ask me anything

Runs Qwen3-0.6B on your device. Code blocks appear as independent cards. Label code with filenames to save & preview.

Settings

Conversations

AI Model

Switching models reloads the page. Already-downloaded models are cached on device — switching back is instant.

Storage

Cached models Calculating…

Deletes all downloaded models from your device. You'll re-download on next use.

Context

Add facts, preferences, or background info. The AI retrieves the most relevant notes using BM25 and factors them in automatically.

Export / Import

Export saves the full conversation to a JSON file. Import restores a previous session.