ChatPing: AI Chat & Assistant

Run a private AI on your phone. ChatPing downloads Gemma, Qwen or Llama to your device so you can chat offline, with no internet connection and no usage caps. Your questions, drafts and documents stay on the phone in your hand. Want a bigger brain for a hard question? Add your own OpenRouter key and use cloud models too, as the second option rather than the first. No account, no sign-up, no cloud sync.

ON-DEVICE MODELS: PRIVATE AND OFFLINE
• Choose from four models that ship in the app: Gemma 4 E2B, Qwen3.5 2B, Qwen3.5 4B and Llama 3.2 3B.
• Download a model once, keep it, and use it anywhere: on a plane, on the subway, in the mountains, or with mobile data switched off.
• When you chat with an on-device model, nothing is sent anywhere. Your messages and replies never leave the device.
• Models range from about 1.3 GB to 3.1 GB, so pick a smaller one to save space or a larger one for richer answers. Downloads run over Wi-Fi, and you can see progress while they finish.
• Switch between your downloaded models any time from the model picker, and start a new chat with a different one when you want to compare answers.
• If an on-device answer isn't enough, retry the same question in the cloud with one tap.

ATTACHMENTS: EVERY TYPE, EVERY MODEL
• Attach photos, PDFs and text documents to a message and ask about them, using an on-device model, entirely offline.
• Works with every model, whether or not the model can see images. Models that can't see pictures still read the text inside your photos, so a receipt, a sign or a screenshot of a document is no problem.
• Summarise a long PDF, then keep asking follow-up questions about it. Pull out the dates, the amounts, the action items or the one clause you were worried about.
• Attach a note, a log or a draft as a text file and ask for a rewrite, a summary or a list of problems.
• Your files are read on your phone. With an on-device model they are never uploaded.

UNLIMITED CHATS, NO LIMITS
• Start as many chats as you like and keep each one going as long as you like. There are no message caps on on-device models.
• Every chat is saved with an automatic title, so you can find it again later, and your history is saved on your phone.
• Search your history by words, open any chat to pick up where you left off, and select several chats at once to delete them.
• Use one chat per project, per trip, per assignment or per idea, and never worry about running out.

CLOUD MODELS WITH YOUR OWN KEY
• Add your OpenRouter key once. It is stored on your device.
• Pick from GPT, Claude, Gemini, Llama, Mistral, DeepSeek, Qwen and more, or enter any custom model ID, and pay only for what you use, directly to OpenRouter.
• Cloud chats go to the model provider you pick, through your own key. ChatPing has no account system and does not sit in the middle.
• Keep a private on-device model for everyday questions and reach for a cloud model only when you need one.

EVERYDAY CHAT THAT FEELS GOOD
• Replies stream in live as they are written, with clean markdown formatting, code blocks and a copy button for replies and code.
• Type your message or dictate it by voice.
• Ask ChatPing to search past chats, open one, switch model or delete a chat for you. It always asks you to confirm before deleting anything.
• Made for phones: a quiet, focused chat screen, light and dark themes, and a history you can browse in a moment.

WHO IT'S FOR
• Anyone who wants an AI assistant that works without a signal and without handing their conversations to a server.
• Students, writers, travellers and professionals who work with private notes, documents and photos.
• Developers and tinkerers who want to run local models on a real phone.
• People who like to try different open models, such as Gemma, Qwen and Llama, side by side.

Privacy policy: https://pingappslab.com/chatping/privacy/
Terms of use: https://pingappslab.com/chatping/terms/

Not affiliated with OpenRouter, Google, Meta, Alibaba or any model provider.