PrivAI LLM - Local AI Chat

PrivAI runs AI language models entirely on your iPhone. Download a model once, then chat offline — nothing you type is sent to a server.

WHAT'S NEW IN 1.5
• Multi-turn conversations — follow-up questions keep their context
• New models: Qwen3.5, Qwen3, Gemma 3, Llama 3.2
• Deep-thinking mode (Qwen3 / Qwen3.5) with a collapsible reasoning trail
• Regenerate, copy and stop at any time; set your own system prompt
• Available in English, 简体中文, 繁體中文 and 日本語

FEATURES
Private by design — conversations are processed and stored on your device. No account, no cloud.

Works offline — after the one-time model download, no internet connection is needed.

Choose your model — the list shows each model's size and whether your device has enough memory. Advanced users can load any MLX model by its Hugging Face ID.

History — search, star and reuse past prompts.

Prompt templates — translate, summarize, polish writing, draft emails.

Photo to text — capture text from paper documents with the camera.

Download mirror — set a custom mirror URL if Hugging Face is slow in your region.

GOOD TO KNOW
Models are about 0.4–3 GB, so Wi-Fi is recommended for the first download. Larger models need more memory; if the app closes while generating, pick a smaller model. AI output may be inaccurate — check important information.