Pipit: Building a 100% On-Device AI Universal Search Engine for Android (<50ms, Open Source)

Description and use

Pipit is an open‑source, on‑device AI search engine for Android that indexes PDFs, receipts, screenshots, gallery, messages, contacts, video, audio, and apps, delivering sub‑50 ms latency without cloud dependence.

Problem

Finding content on smartphones is difficult because stock search only checks filenames and cloud search uploads private files.

Outcome

Pipit delivers a unified, offline search across seven domains with sub‑50 ms latency, eliminating the need for cloud uploads.

Workflow

• Index local media using Google ML Kit OCR for PDFs and receipts
• Extract visual features for semantic image search
• Store embeddings in a local database
• Query unified search bar with natural language
• Return results from all domains within 50 ms

Evidence

Smartphones hold our entire digital lives: thousands of photos, receipts, work PDFs, voice notes, and gate codes buried in messages.

Source

Yet finding anything is painful because stock search only checks cryptic filenames like IMG_20240914.jpg or Scan_Doc_09.pdf.

Source

Pipit 🐦 — an open-source, on-device AI universal search engine for Android.

Source

Deep Content OCR: Searches inside PDFs, receipts, and screenshots using on-device Google ML Kit.

Source

Semantic Visual Search: Natural language search across your gallery ("yellow vintage car", "handwritten grocery list").

Source

Unified Search Bar: 7 domains in one place (Photos, PDFs, Messages, Contacts, Video, Audio, Apps).

Source

100% Offline & Air-Gapped: Runs quantized Google LiteRT multimodal models directly on your phone's silicon. Zero cloud, zero tracking.

Source

Sub-50ms Latency: Accelerated via local Vulkan GPU/NPU delegates.

Source

Source code and release APKs are available on GitHub: Repository: https://github.com/sanjay434343/Pipit

Source