# Offline AI - Private Chat — Local LLM, PDF & Image Search > On-device AI chat with private, offline model execution and customizable features. (iPhone/iPad app by Twin Suns LLC.) - Source: https://appshunter.io/ios/app/offline-ai-private-chat/id6760442643 (this page in markdown: same URL + `.md`) - Developer: [Twin Suns LLC](https://appshunter.io/developer/1834800642) - Category: Utilities - Price: Free with in-app purchases - Rating: 4.10/5 from 12 App Store ratings · 2 written reviews indexed - Age rating: 17+ - Requires: iOS 18.0 · 52 MB - Languages: American English - Released: 2026-03-14 - Data updated: 2026-09-05 - Monetization: freemium - User reviews in markdown: https://appshunter.io/ios/app/offline-ai-private-chat/id6760442643/reviews.md ## What is Offline AI? Run powerful AI models directly on your iPhone — no internet required, no data ever leaves your device. PRIVATE BY DESIGN Every response is generated on your device using Apple's Metal GPU. No API keys. No cloud. No tracking. No account. ASK ACROSS YOUR DOCUMENTS Save contracts, manuals or notes to your library, then ask about them from any chat. The passages that answer your question are found for you, so a long PDF no longer has to be cut short. Every answer shows which file it came from. CREATE IMAGES Describe a picture and it's generated on your device. Nothing you type or make is uploaded. SEE AND SPEAK Vision models read text in photos and describe what they see. Voice mode reads replies aloud, hands-free. BROWSE & DOWNLOAD MODELS Explore curated categories — Reasoning, Coding, General and Creative — filtered to fit your device's RAM. Search HuggingFace live to find any model. CHAT MODES & PERSONAS Seven response modes — LaTeX, Code, Concise, Step-by-step, ELI5, Formal and Email Reply. Give your AI a custom Persona with a name, icon and system prompt. WEB SEARCH Research Mode searches the web before responding, so any model can answer about current events. Off by default. STAY INFORMED The Discover tab surfaces the latest AI news, updated through the day. OPTIMIZED FOR IPHONE Best results on iPhone 15 Pro and later with 8GB RAM. Pro unlocks voice mode, model arena, vision models and image generation. Subscriptions renew automatically unless cancelled at least 24 hours before the end of the period. Terms of Use (EULA): https://www.apple.com/legal/internet-services/itunes/dev/stdeula/ Privacy Policy: https://myai.twinsuns.dev/privacy ## Key features - On-device AI model execution - Complete offline functionality - Private by design, no data leaves device - Browse and download AI models - Customizable chat modes and personas - Integrated web search for current events - AI news and trending model discovery ## Pricing and in-app purchases Base price: Free. **Subscriptions** | Subscription | Price | Period | | --- | --- | --- | | Pro Monthly | $4.99 | per month | | Pro Annual | $29.99 | per year | **One-time purchases** | Purchase | Price | Type | | --- | --- | --- | | Pro Lifetime | $39.99 | one-time | ## Recent user reviews (2 of 2) All indexed reviews: https://appshunter.io/ios/app/offline-ai-private-chat/id6760442643/reviews.md ### 5/5 — Many Features *2026-08-16* This app and new features are constantly being pushed ### 1/5 — Can't upload my own images *2026-08-08* Getting a StableDiffusion.Encoder.Error error 0.) **Developer response:** You were right, and the error text you included is exactly what let me find it — thank you. This was my bug, not your device. Reference images weren't being resized to the dimensions the model expects, so any photo from your library failed on import. It affected everyone using image-to-image. It's fixed in 1.17, which is approved and live now. Update and your own images will import properly. Sorry you ran into it. If anything else gives you trouble, myai.twinsuns.dev/support comes straight to me. ## Frequently asked questions about Offline AI ### Does Offline AI have ads? Offline AI is ad-free, allowing for an uninterrupted user experience. The app is free to download and use with core features. ### What are the subscription options for Offline AI Pro? Offline AI Pro offers monthly ($4.99/month), annual ($29.99/year), and a lifetime purchase ($59.99) for advanced features. Subscriptions auto-renew unless cancelled 24 hours prior to the end of the current period. ### How does Offline AI ensure privacy? Offline AI is designed for privacy by keeping all conversations and AI model processing on your device. No API keys, cloud services, or tracking are used, ensuring your data never leaves your iPhone. ### What devices are recommended for Offline AI? For the best results, Offline AI is optimized for iPhone 15 Pro and later models with 8GB of RAM. It is also supported on iPads. ### How often is Offline AI updated? The latest version of Offline AI is 1.12, last updated on May 17, 2026. Updates typically include new features, model optimizations, and bug fixes. ### Can I use Offline AI without an internet connection? Yes, a key feature of Offline AI is its ability to run powerful AI models entirely on your device without requiring an internet connection. This ensures complete offline functionality. ### What advanced features are available with Offline AI Pro? Offline AI Pro unlocks Arena Mode for side-by-side model comparison, Voice Mode with premium TTS, HuggingFace Live Search for any GGUF model, Vision Models, on-device Image Generation with Stable Diffusion, and unlimited saved models. ### What is the age rating for Offline AI? Offline AI has an age rating of 17+, indicating it may contain content suitable for mature audiences. This is common for AI applications that can generate a wide range of text and creative outputs. ## Version history (last 5 releases) ### 1.19 — 2026-08-19 Search your chats The chat list now has a search bar. It looks inside your conversations, not just their titles, so you can find a chat by something that was said in it. Results show the matching text, so you can tell which one you meant without opening each. ### 1.17 — 2026-08-15 Document Library (Pro) Save documents once and ask questions across all of them from any chat. Turn on "Search my library" and the relevant passages are found and given to the model, rather than the document being cut short. A long PDF used to be trimmed to a fraction of its length, with the model answering as though it had read the rest. Length is no longer the limit. Answers that used your documents show which file each excerpt came from, the passage itself, and how closely it matched — so you can check it pulled from the right place. You can also narrow a chat to specific documents when you already know which one holds the answer. Documents are read, indexed and stored on your device. Nothing is uploaded. Fixed Using one of your own photos as a starting point for image generation failed with an encoder error. Photos are now prepared at the size the model expects, and a portrait photo no longer comes out rotated. Also The tokens-per-second readout under each reply can be turned off in Profile. It was only ever measured on your device and never recorded or sent anywhere. ### 1.16 — 2026-08-14 Fixed a bug where a long conversation could stop responding entirely. Once it hit the context limit, every following message failed the same way with no way to recover short of starting a new chat. Older messages are now set aside automatically so the conversation keeps going. Models also reserve memory for context more efficiently, which leaves more room on smaller devices, and attached documents are now sized to the model you have loaded instead of a fixed limit. Added Report a Bug in Profile — it drafts an email with your app version and device already filled in. ### 1.15 — 2026-08-06 LaTeX mode writes maths, not documents Asking for a formula in LaTeX mode used to return an entire compilable document — preamble, packages and all — which arrived as raw source instead of a rendered equation. It now writes equations inline with the explanation, the way you'd expect, and keeps explaining in plain language around them. Honest web search DuckDuckGo intermittently blocks automated searches. Until now that looked exactly like finding nothing, so the answer quietly came from the model's own knowledge while appearing to have been freshly searched. The app now tells you when a search didn't run, and the model says so too, rather than presenting a confident answer as though it were current. Sources stay with their answer Search citations now sit directly beneath the reply they produced, and survive reopening the conversation. Previously they always showed the most recent search and were replaced the moment you asked something else — scrolling back to an earlier answer showed sources belonging to a later question, or none at all. Smaller download Removed 1.4 MB of unused resources left over from an older maths renderer. ### 1.14 — 2026-08-04 Know what will run before you download it Every model now shows whether it runs comfortably on your iPhone, runs but uses most of your memory, or won't fit at all — measured against your device's actual RAM. The same badge appears in Discover, so you can tell at a glance instead of spending four gigabytes finding out. Newer models The on-device inference engine is roughly 1,900 builds newer. Qwen3.5 and other recent architectures now load and run properly instead of failing partway through. Reasoning models, properly handled Models that think before answering now collapse that thinking instead of showing it as the answer. Their reasoning no longer fiher, so long conversations stay coherent. And the response-length setting now limits the answer rather than the thinking — previously and its whole budget deliberating and never reach a reply. Formatting that holds together Headings, lists and bold text render as intended rather than leaking raw ** markers into replies, and line breaks stay whekdown now always renders, so the Markdown chat mode has been removed — it was a switch that appeared to do nothing. Fenced code appears as. Fixes • Em-dashes, accented characters and emoji courd • Headlines in Discover showed ’ instead of an apostrophe • Deleting a model while it was loading could • Models that failed to load did so silently — they now tell you why • A model could appear in your library with no • Some models were listed twice with conflicting memory requirements ## More apps by Twin Suns LLC - [Level Up AI: Soccer Training](https://appshunter.io/ios/app/level-up-ai-soccer-training/id6751371008) - [Huma Create](https://appshunter.io/ios/app/huma-create/id6756941315) - [SideQuest: AI Goal Quests](https://appshunter.io/ios/app/sidequest-ai-goal-quests/id6757112707) - [Cookfolio AI](https://appshunter.io/ios/app/cookfolio-ai/id6757167426) - [Baseline: Blood Test Tracker](https://appshunter.io/ios/app/baseline-blood-test-tracker/id6760978817) All apps by Twin Suns LLC: https://appshunter.io/developer/1834800642 ## Related topics [uncensored ai](https://appshunter.io/ios/topics/uncensored-ai) --- *Data collected daily from the US App Store and indexed by [AppsHunter](https://appshunter.io/). User reviews are verbatim App Store reviews. Ratings, prices and chart positions refresh continuously; this snapshot is from 2026-09-05.*