a-ghorbani /pocketpal-ai
An app that brings language models directly to your phone.
★ 9,352⑂ 1,105👁 73⑃ 472● 544▤ 95
+60
Daily
9,352
stars
1,105
forks
73
watchers
472
prs
544
issues
95
releases
README
PocketPal AI A private AI assistant that runs entirely on your phone. Chat with language models, give them a voice, and let them use tools — all on-device. No account, no cloud, no internet required. pocketpal.dev · Get the app · Leaderboard · PalsHub · Discussions --- Why PocketPal AI? Most AI apps are a thin window onto someone else's server — every message you type gets shipped off, logged, and analyzed somewhere you can't see. PocketPal flips that around: the AI lives on your phone, and your conversations stay on it unless you choose an online feature. - 🔒 Private by default — every prompt, response, and document stays on your device unless you turn on an online feature such as web search or a remote server. - ✈️ Works offline — download a model once and it just works, with no connection and no account. On a plane, on a trail, anywhere. - 📱 Runs on hardware you already own — real language models, voices, and tools, tuned to make the most of your phone's CPU, GPU, and NPU. - 🆓 Free and open source — chat, models, voices and tools are free, with no account. The App Store and Google Play builds offer an optional one-time PocketPal Plus purchase for hands-free voice and the local API server; builds from source include it, and so does the APK on GitHub Releases from the release after v1.23.0. MIT-licensed. Privacy note: Your chats and documents leave your device only through features you turn on: web search and page reading, a remote server, or a system voice your phone runs online. The app's other network use (model and voice downloads, PalsHub and purchases, device-compatibility rules from the jsDelivr CDN at launch, the opt-in benchmark leaderboard, and feedback or reports you submit) never attaches your chats. Contents - Features - Get the app - How it works - Using the app - For developers - Contributing - Roadmap - Community & support - License Features - 🧠 On-device chat — run GGUF language models (Gemma, Qwen, Phi, Llama, and more) fully offline. - 🗣️ Text-to-speech — give your assistant a voice with on-device neural voices (Kokoro, Kitten, Supertonic, and Phonon on supported devices) that synthesize speech on your phone, or your phone's system voices. - 🎙️ Voice input — dictate messages with on-device Whisper speech recognition (whisper.cpp). - 🎭 Pals — create personalized assistants with their own model, system prompt, and personality (Assistant and Roleplay types). - 🛍️ PalsHub — discover and install community Pals, including premium ones via in-app purchase. - ✨ PocketPal Plus — an optional one-time purchase in the App Store and Google Play builds that adds hands-free voice conversation and a local API server. Builds from source, and the APK on GitHub Releases from the release after v1.23.0, include every Plus feature with no purchase. - 🛠️ Talents & tools — let capable Pals call built-in tools (calculator, date/time, rich HTML rendering, opt-in web search and page reading) inside a tool-use loop. - 📥 Huggi
Top issues
#55[General]: Publish in FDroid21 comments · 2024-10-21#494[General]: Cannot import .guff file from "Add local Model" function11 comments · 2025-11-29#143[Build] Separate firebase-enabled and non-firebase builds11 comments · 2024-12-21#183[Feat]: add files in chat6 comments · 2025-01-26#389[Bug]: When using multimodal AI, the + button is always inactive.5 comments · 2025-08-03#306[Feat]: Support for LiteRT (aka TensorFlow Lite or TFLite) for a 10x performance increase5 comments · 2025-05-11#401[Feat]: Add ability to attach text documents + automatic RAG activation when context limit is exceeded4 comments · 2025-08-09#138[Feat]: Add support for LaTeX formatting in PocketPal's.4 comments · 2024-12-12
Discussions
#1031Marco-Nano 1- to 3-bit GGUFs and a Pal: 25 to 37 tokens/s on a Pixel 10a0 comments · 2026-10-28#985Ability to adjust line spacing in chat0 comments · 2026-10-16#984Execute tool calls concurrently to improve user experience1 comments · 2026-10-16#958Model Output Errors with Tested Models2 comments · 2026-10-06#946App to Web API2 comments · 2026-10-02#918Ralph Device Lab: a 2.94 GB Qwen3-8B GGUF on four physical iPhones (PocketPal receipts)0 comments · 2026-09-15#898Much slower performance after latest version0 comments · 2026-09-04#889A tuning-layer fork contribution intent: OpenCL backend fixes inbound (Mali fp16 GEMM, F16 KQ/KQV guard)0 comments · 2026-09-02
Latest release
Release v1.23.0 v1.23.0 2026-11-10What's Changed [Bug]: iOS chat input stays tall after sending a multi-line message by @pocketpal-dev-team[bot] in https://github.com/a-ghorbani/pocketpal-ai/pull/1052 [Bug]: Search download over a text-only preset skips vision by @pocketpal-dev-team[bot] in https://github.com/a-ghorbani/pocketpal-ai/pull/1053 fix(e2e): don't report runs where every test skipped as passed by @alicangunduz in https://github.com/a-ghorbani/pocketpal-ai/pull/1054 [Feat]: Local OpenAI-compatible API server (POC) by @pocketpal-dev-team[bot] in https://github.com/a-ghorbani/pocketpal-ai/pull/990 [Bug]: Imported