Atomic Chat
Free, open-source local AI chat for desktop and mobile with offline models, private chats, local APIs, and agent workflows.
🛡️ AgentReady threat assessment
MAESTRO 7-layer threat model + OWASP AIVSS risk score for Atomic Chat, derived from its capabilities.
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology — every score is re-derived by the same automated method as an agent's public evidence changes.
Overview
Atomic Chat is a free, open-source local AI chat app for Mac, Windows, Linux, iOS, and Android. It runs models such as Llama, Qwen, DeepSeek, Mistral, Gemma, Kimi, and MiniMax locally, with support for browsing and downloading 1,000+ models from Hugging Face. The official site describes Atomic Chat as private by design: no cloud requirement, no subscription, no rate limits, and data stays on the user’s device. It also includes agent-focused capabilities for creating and running autonomous workflows locally, plus local API support, chats and projects, persistent memory across sessions, artifacts, PDF chat, and model formats including GGUF, MLX, and ONNX.
Key features and capabilities
- local ai
- offline chat
- privacy
- open source
- llama
- qwen
- deepseek
- gemma
- hugging face
- local api
Use cases
- Running private AI chats locally without sending data to the cloud
- Using open-source models offline on desktop or mobile devices
- Creating and running local autonomous workflows on your own machine
- Organizing chats and projects with persistent memory across sessions
- Downloading and testing models from Hugging Face with one-click setup