Off Grid
Run AI text, image gen, vision offline on your phone
Details
- External ID
- 47019133
- Source
- HN
- Company
- —
- Product
- Off Grid
- Website domain
- github.com
- Launched
- Feb. 14, 2026
- Cohort
- —
- Upvotes
- 124
- Upvotes percentile
- 0.9110512129380054
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:25 p.m.
- Updated at
- Sept. 7, 2026, 9:25 p.m.
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.
Enrichment
- Theme
- voice dictation and control tools
- Vertical
- Horizontal
- Function
- Model & infra
- Audience
- B2C
- AI stance
- AI-native
- Project type
- Commercial product
- Normalized one-liner
- offline ai text and image generation on mobile
- Manually corrected
- False
Could you build this?
Partial The UI and high-level mobile shell are standard, but running multi-modal LLMs, vision, and diffusion models locally on mobile GPUs requires low-level hardware acceleration and quantization tuning.
What it would actually take: The app needs to integrate low-level inference backends such as llama.cpp / MLC-LLM / Core ML / Android NNAPI with tailored GPU shader backends (Metal / Vulkan). Key challenges include memory paging and thermal throttling management on mobile devices, compiling custom GGML/GGUF quantization kernels, and optimizing diffusion models for severely constrained unified RAM pools.
Discussion
20 comments analyzed.
Competitors mentioned: Russet for iOS/iPadOS, FUTO keyboard, Google AI Edge Gallery, Ollama + OpenWebUI
Concerns raised: Image generation requires network access despite offline claims, NPU support limited to Qualcomm chips, Text generation speed (15-30 tok/s) slower than alternatives, Small quantized models (4B-8B) lack sufficient context window, App allows downloading models that won't fit in available RAM
Feature requests: Markdown rendering in chat responses, Hardware-based model recommendations before download, Pre-download validation of RAM requirements
Competitors
Other products that read as similar to this one — 46 launches clear the similarity bar, closest 8 shown.
Attention rank: #5 of 47 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 49 days after the earliest competitor.
- Off Grid: On-device AI-web browsing, tools vision,image,voice–3x faster · hn · 2026-02-24 · 12 upvotes · similarity 0.62
- Run AI chat, image gen, vision, and voice offline on your Mac · hn · 2026-06-29 · 11 upvotes · similarity 0.61
- Qwen 3.5 running on a $300 Android phone · hn · 2026-03-03 · 6 upvotes · similarity 0.53
- PocketMind AI · ph · 2026-09-11 · 1 upvotes · similarity 0.47
- AiLocal · ph · 2026-09-28 · 1 upvotes · similarity 0.47
- Offline AI dev assistant (no API, runs locally) · hn · 2026-04-10 · 5 upvotes · similarity 0.40
- LifeContext AI · ph · 2026-09-22 · 2 upvotes · similarity 0.39
- UnslopAI · ph · 2026-09-09 · 1 upvotes · similarity 0.38
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a model & infra tool for Fintech yet.