Atomic Chat
Run open-weight AI models locally on your device, fully private and offline
Visit atomic.chat ↗What is Atomic Chat?
Atomic Chat is a desktop and mobile application that lets you run large language models locally on your machine without cloud dependencies or accounts. It supports over 1,000 open-weight models including Gemma, Qwen, DeepSeek, Llama, and Mistral, with features like speculative decoding for faster inference, Flash Attention, and automatic context-window expansion.
The app includes an OpenAI-compatible API server at localhost:1337/v1, making it work as a drop-in replacement for OpenAI's SDK with any agent, CLI tool, or IDE plugin. Users can also connect to cloud providers like OpenAI, Anthropic, and Groq while keeping local models private. The project is free and open-source.
Momentum
Daily activity over the last 30 days: upvotes, unique visits, new GitHub stars and Hacker News points.
How do we know it's vibe coded?
vibetrend discovered Atomic Chat automatically. Evidence: GitHub AtomicBot-ai/Atomic-Chat; repo has AGENTS.md. View source.
Similar vibe-coded apps
Remote browser infrastructure and agents for AI automation tasks
Multi-account AI tool manager with smart model routing and local LLM deployment
Distribute large language models across multiple machines via one API
Free, open-source AI video generator from text with narration and subtitles
Web UI for AI image generation and editing with multiple API providers
AI prompt generator optimized for Suno music creation across all versions
Communication platform designed for AI agents to interact with each other