# AtomicBot-ai/Atomic-Chat

Local AI app and inference engine for agents. Run open-weight LLMs locally — private, 100% offline on your computer. Join our Discord: https://discord.com/invite/8wGSsvmg4V

Repository: https://github.com/AtomicBot-ai/Atomic-Chat
Canonical: https://ross.abutalabs.com/products/atomic-chat
Homepage: https://atomic.chat
Language: TypeScript
License: NOASSERTION
License Family: other
Topics: ai-chat, ai-tools, apple-silicon, chatgpt, deepseek, desktop-app, gemma, gguf, gpt-oss, llamacpp, llm, llm-inference, local-ai, local-first, local-llm, mcp, mlx, open-source, qwen, self-hosted
Last push: 2026-08-26T14:08:48+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 98, longevity 11
- inputs: {"age_days": 155, "days_push": 7, "days_rel": 12, "gap_med": 1.0, "n_releases_24m": 47}
- flags: young, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1374, forks 158 (observed 2026-08-28T04:04:32.216120+00:00)

## What it is
Atomic Chat is an open-source desktop and mobile application with a built-in inference engine for running open-weight LLMs fully locally and offline. It supports GGUF and MLX models, one-click agent setups, MCP integration, and includes TurboQuant quantization for faster, memory-efficient inference.

## Use cases
- run chatgpt-like ai locally without internet
- chat with local llms privately on my mac
- run deepseek or qwen models offline on windows
- use ai agents without sending data to the cloud
- run open-weight models on iphone or android
- quantized local llm inference with low memory

## When to choose
- you need fully private, 100% offline LLM usage on desktop or mobile
- you want one-click setup for local models and agents with MCP support
- you're on Apple Silicon and want MLX/GGUF inference with quantization speedups

## When to avoid
- you need cloud-scale serving or multi-user deployment
- you rely on the latest proprietary frontier models (GPT-4, Claude)
- you need a permissive open-source license - the license is custom/unverified

## Facets
- artifact type: application
- maturity: active
- function: llm-inference, chatbot, chat-interface, agent-framework, mcp
- domain: large-language-models, artificial-intelligence, self-hosted, privacy
- platform: windows, cross-platform
- tags: local-llm, gguf, llamacpp, mlx, apple-silicon, offline-ai, turboquant, open-weight-models, local-first, ai-agents, macos, linux, ios, android, desktop, mobile, tauri

## Member repositories
- AtomicBot-ai/Atomic-Chat (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:32.216120+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:40:35.917709+00:00, confidence not recorded.
  - readme: https://github.com/AtomicBot-ai/Atomic-Chat (fetched 2026-08-28T04:04:32.216120+00:00, sha 75a0566f9b96)
  - homepage: https://atomic.chat (fetched 2026-08-29T11:57:11.048394+00:00, sha ad3ddedbd28a)
- Data as of 2026-08-30T08:39:29.467469+00:00.