Atomic-Chat
Run open-weight LLMs locally with offline support and OpenAI API compatibility.
- Region
- Overseas
- Pricing
- Free
- Open source
- Yes
- GitHub Stars
- ★ 1.1k
- Source
- GitHub
- Added
- 2026-07-18
- Last verified
- 2026-07-18

Overview
Atomic Chat is a local AI application and inference engine that privately runs open-weight large language models on your computer. It supports multiple local models such as Llama, Gemma, Qwen, and provides various optimization techniques to enhance performance. Additionally, Atomic Chat includes several cloud model providers and allows users to freely switch between local and cloud models. The app also features one-click startup of coding agents and real-time code preview, making it ideal for developers and teams.
Key features
- ▪Run open-weight LLMs locally
- ▪Supports multi-model optimization techniques
- ▪OpenAI API compatible
- ▪Built-in multiple cloud model providers
- ▪One-click startup of coding agent
Use cases
Pros
- +Fully offline usage
- +Supports multiple local models
- +Strong API compatibility
- +Rich tool integration
Limitations / notes
- -Requires some configuration
- -Some features require internet connection
Who it's for
This overview was compiled by AI from public sources and may contain inaccuracies — please refer to the official site.
FAQ
Is it free?
Yes, Atomic Chat is free and open-source.
Does it support Chinese?
Yes, it can run local models that support Chinese.
Can it be used commercially?
Yes, but you must comply with the relevant open-source licenses.
Does it require internet connection?
Local usage does not require internet, but some features may need it.
Similar agents
Cursor
FeaturedAI-powered code editor with Agent mode
Dify
FeaturedOpen-source LLMOps + Agent orchestration platform
扣子 Coze
FeaturedByteDance's no-code Agent/Bot building platform
Trae
FeaturedAI IDE (with Agent) from ByteDance
LangChain
FeaturedThe most mainstream LLM/Agent orchestration framework
Devin
FeaturedAutonomous software engineer Agent by Cognition
Something wrong? Let us know on the About page and we'll fix it.