GitHub Repository

The Swiss Army Knife of Offline AI. Chat, Speak, and Generate Images - Privacy First, Zero Internet. Download an LLM and use it on your mobile device. No data ever leaves your phone. Supports text-to-text, vision, text-to-image

2,346 starsTypeScript

Off Grid – Run AI text, image gen, vision offline on your phone

Name: Off Grid – Run AI text, image gen, vision offline on your phone
Availability: InStock
Author: ali_chherawalla

by ali_chherawalla·Feb 14, 2026·124 points·66 comments

Visit Project View on HN

AI Analysis

●●●BangerZero to OneSolve My ProblemWizardry

Complete offline AI suite on mobile: text, images, vision, voice—no servers, no uploads, real speeds.

Strengths

•Genuinely solves privacy + accessibility gap: offline AI on consumer hardware with measurable performance (15-30 tok/s text, 5-10s image gen).
•Hardware acceleration across platforms (Snapdragon NPU, Core ML, Metal) shows serious mobile engineering—not just wrapping an LLM.
•Multi-modal (text, vision, image gen, voice) in one cohesive app—goes far beyond single-task local LLM competitors.

Weaknesses

•Model downloads and storage still friction; unclear how seamless onboarding is for non-technical users despite GIFs.
•Performance gap between flagship and mid-range devices may limit real-world usability for majority of smartphone users.

Post Description

Your phone has a GPU more powerful than most 2018 laptops. Right now it sits idle while you pay monthly subscriptions to run AI on someone else's server, sending your conversations, your photos, your voice to companies whose privacy policy you've never read. Off Grid is an open-source app that puts that hardware to work. Text generation, image generation, vision AI, voice transcription — all running on your phone, all offline, nothing ever uploaded.

That means you can use AI on a flight with no wifi. In a country with internet censorship. In a hospital where cloud services are a compliance nightmare. Or just because you'd rather not have your journal entries sitting in someone's training data.

The tech: llama.cpp for text (15-30 tok/s, any GGUF model), Stable Diffusion for images (5-10s on Snapdragon NPU), Whisper for voice, SmolVLM/Qwen3-VL for vision. Hardware-accelerated on both Android (QNN, OpenCL) and iOS (Core ML, ANE, Metal).

MIT licensed. Android APK on GitHub Releases. Build from source for iOS.