Atomic-Chat
<img src="https://github.com/AtomicBot-ai/Atomic-Chat/raw/main/assets/logo.png" width="80" alt="Atomic Chat" /> # Atomic Chat Local AI app and inference engine for agents. Run open-weight LLMs locally — private, on your machine. <a href="https://github.com/AtomicBot-ai/Atomic-Chat/stargazers"><img src="https://img.shields.io/github/stars/AtomicBot-ai/Atomic-Chat?style=flat&logo=github&label=Stars&color=f5c542" alt="Stars" /></a> <a href="https://github.com/AtomicBot-ai/Atomic-Chat/network/members"><img src="https://img.shields.io/github/forks/AtomicBot-ai/Atomic-Chat?style=flat&logo=github&label=Forks&color=4ac1f2" alt="Forks" /></a> <a href="https://github.com/AtomicBot-ai/Atomic-Chat/graphs/contributors"><img src="https://img.shields.io/badge/Contributors-10-ff69b4?style=flat&logo=github" alt="Contributors" /></a> <a href="https://github.com/AtomicBot-ai/Atomic-Chat/commits/main"><img src="https://img.shields.io/github/last-commit/AtomicBot-ai/Atomic-Chat?style=flat&label=Last%20Commit&color=blueviolet" alt="Last Commit" /></a> <img src="https://img.shields.io/badge/Built_with-Tauri-FFC131?style=flat&logo=tauri&logoColor=white" alt="Tauri" /> <img src="https://img.shields.io/badge/Runtime-Node.js_≥20-339933?style=flat&logo=nodedotjs&logoColor=white" alt="Node.js" /> [Getting Started](https://atomic.chat/) · [Discord](https://discord.com/invite/8wGSsvmg4V) · [X / Twitter](https://x.com/atomic_chat_hq) · [Bug Reports](https://github.com/AtomicBot-ai/Atomic-Chat/issues) <p align="center"> <img src="https://github.com/AtomicBot-ai/.github/raw/main/assets/0529_final.gif" width="100%" alt="Atomic Chat — local AI chat in action" /> </p> --- ### Download <p align="left"><b>Desktop</b></p> <p align="left"> <a href="https://github.com/AtomicBot-ai/Atomic-Chat/releases/download/v1.1.95/Atomic.Chat_1.1.95_universal.dmg"><img src="https://img.shields.io/badge/macOS-Universal-000000?style=for-the-badge&logo=apple&logoColor=white" height="46" alt="Download for macOS" /></a> <a href="https://github.com/AtomicBot-ai/Atomic-Chat/releases/download/v1.1.95/Atomic.Chat_1.1.95_x64-setup.exe"><img src="https://img.shields.io/badge/Windows-x64-0078D4?style=for-the-badge&logo=windows11&logoColor=white" height="46" alt="Download for Windows" /></a> <a href="https://github.com/AtomicBot-ai/Atomic-Chat/releases/download/v1.1.95/Atomic.Chat_1.1.95_amd64.AppImage"><img src="https://img.shields.io/badge/Linux-AppImage-FCC624?style=for-the-badge&logo=linux&logoColor=black" height="46" alt="Download for Linux" /></a> </p> <p align="left"><b>Mobile</b></p> <p align="left"> <a href="https://apps.apple.com/us/app/atomic-chat-private-local-ai/id6761720226"><img src="https://img.shields.io/badge/iOS-App_Store-0D96F6?style=for-the-badge&logo=appstore&logoColor=white" height="46" alt="Download for iOS" /></a> <a href="https://play.google.com/store/apps/details?id=chat.atomic.app"><img src="https://img.shields.io/badge/Android-Google_Play-414141?style=for-the-badge&logo=googleplay&logoColor=white" height="46" alt="Download for Android" /></a> </p> or grab any build from <a href="https://atomic.chat/">atomic.chat</a> · <a href="https://github.com/AtomicBot-ai/Atomic-Chat/releases">GitHub Releases</a> — latest: <a href="https://github.com/AtomicBot-ai/Atomic-Chat/releases/tag/v1.1.95"><b>v1.1.95</b></a> --- ### Contributors Atomic Chat is built by a small team and a handful of community contributors. Pull requests welcome — see [CONTRIBUTING.md](CONTRIBUTING.md) for how to get started. <a href="https://github.com/Vect0rM"><img src="https://images.weserv.nl/?url=https://github.com/Vect0rM.png&w=120&h=120&fit=cover&mask=circle" width="60" height="60" alt="Vect0rM" /></a> <a href="https://github.com/dtorey-d"><img src="https://images.weserv.nl/?url=https://github.com/dtorey-d.png&w=120&h=120&fit=cover&mask=circle" width="60" height="60" alt="dtorey-d" /></a> <a href="https://github.com/MaxKoshJob"><img src="https://images.weserv.nl/?url=https://github.com/MaxKoshJob.png&w=120&h=120&fit=cover&mask=circle" width="60" height="60" alt="MaxKoshJob" /></a> <a href="https://github.com/Albert-Atomic"><img src="https://images.weserv.nl/?url=https://github.com/Albert-Atomic.png&w=120&h=120&fit=cover&mask=circle" width="60" height="60" alt="Albert-Atomic" /></a> <a href="https://github.com/yanalialiuk"><img src="https://images.weserv.nl/?url=https://github.com/yanalialiuk.png&w=120&h=120&fit=cover&mask=circle" width="60" height="60" alt="yanalialiuk" /></a> <a href="https://github.com/corevibe555"><img src="https://images.weserv.nl/?url=https://github.com/corevibe555.png&w=120&h=120&fit=cover&mask=circle" width="60" height="60" alt="corevibe555" /></a> <a href="https://github.com/claytonlin1110"><img src="https://images.weserv.nl/?url=https://github.com/claytonlin1110.png&w=120&h=120&fit=cover&mask=circle" width="60" height="60" alt="claytonlin1110" /></a> <a href="https://github.com/Fieldnote-Echo"><img src="https://images.weserv.nl/?url=https://github.com/Fieldnote-Echo.png&w=120&h=120&fit=cover&mask=circle" width="60" height="60" alt="Fieldnote-Echo" /></a> <a href="https://github.com/Angelopgit"><img src="https://images.weserv.nl/?url=https://github.com/Angelopgit.png&w=120&h=120&fit=cover&mask=circle" width="60" height="60" alt="Angelopgit" /></a> --- ### Features **Local models** - Run open-weight LLMs locally from HuggingFace — Llama, Gemma, Qwen, Mistral, Phi, and others - Multi-Token Prediction (MTP) speculative decoding — 30–70% throughput boost on supported models, up to 3× on Gemma 4 - DFlash block-diffusion decoding — up to 6× faster on Qwen 3.6, Gemma 4, Kimi K2.5 - Flash Attention toggle (`on` / `off` / `auto`) - Automatic reasoning-context tracking for chain-of-thought models - Auto context-window expansion with overflow notifications - EAGLE-3 speculative decoding for Gemma 4 on Apple Silicon (MLX) - MTP on MLX for Qwen 3.5 / 3.6 and DeepSeek V4 - TurboQuant KV cache on MLX-VLM — smaller memory footprint via RHT-correct fast paths **Cloud models** - Built-in providers: OpenAI, Anthropic, Mistral, Groq, MiniMax, Qwen, Moonshot - Bring your own key, switch model per chat, mix local and cloud freely **Tools & integrations** - One-click agent launch — launch OpenCode and GitHub Copilot CLI agents in one click from the Integrations tab - Artifacts — live preview panel for HTML/CSS/JS code with copy, download and print - Connect multiple [MCP](https://modelcontextprotocol.io/) servers — bring your own tools, file access, web search - In-app log viewer for MCP tool calls - Custom assistants with per-assistant system prompts - Projects with conversation tree view in the sidebar **Local API** - OpenAI-compatible server at `http://localhost:1337/v1` — drop-in replacement for the OpenAI SDK - Works with any agent, CLI, or IDE plugin that speaks the OpenAI API - Bound to `127.0.0.1` by default; set `host: 0.0.0.0` to expose on LAN **Privacy** - Everything runs locally when you want it to — local server is loopback-only by default - Your conversations and keys stay on your machine --- ### Inference Engines Three engines under the hood, all exposed through one OpenAI-compatible API at `http://localhost:1337/v1`: - **[atomic-llama-cpp-turboquant](https://github.com/AtomicBot-ai/atomic-llama-cpp-turboquant)** — our `llama.cpp` fork with TurboQuant optimizations for faster quantized inference. Cross-platform (macOS, Windows, Linux), CPU and GPU. - **Upstream [llama.cpp](https://github.com/ggml-org/llama.cpp)** — official `ggml-org` build, used on Windows by default for the widest hardware coverage and MTP support. - **[MLX-VLM](https://github.com/Blaizzy/mlx-vlm)** — Apple Silicon-native engine for vision-language models, running on the Neural Engine and unified memory. Faster than llama.cpp on M-series chips for supported models. Speculative-decoding features available across backends: - **MTP (Multi-Token Prediction)** — a draft model predicts ahead, the full model verifies in one pass. Available on macOS and Windows. - **DFlash** — block-diffusion speculative decoding for Qwen 3.6, Gemma 4, Kimi K2.5 and others. Apple Silicon only; can't be enabled together with MTP. - **Flash Attention** — Settings → `on` / `off` / `auto`. Tools talking to `http://localhost:1337/v1` don't need to know which backend is running underneath — switch engines without reconfiguring clients. --- ### Launch With Atomic Chat runs an OpenAI-compatible server at `http://localhost:1337/v1`, so **any agent, CLI, IDE plugin, or app that speaks the OpenAI API can run on top of your local models** — no extra glue needed. Just point its base URL at Atomic Chat and you're done. A few projects already ship first-class support with their own setup docs: | Tool | What it is | Setup | | --- | --- | --- | | **[OpenCode](https://opencode.ai/)** | Open-source TUI coding agent. Add Atomic Chat as a local provider in `opencode.json`. | [Setup guide →](https://opencode.ai/docs/providers/#atomic-chat) | | **[Goose](https://github.com/block/goose)** | Open-source extensible AI agent (CLI, desktop, API). | [Setup guide →](https://goose-docs.ai/docs/getting-started/providers/#local-llms) | | **[nanobot](https://github.com/HKUDS/nanobot)** | Ultra-lightweight personal AI agent with chat channels, MCP, and WebUI. | [Repo →](https://github.com/HKUDS/nanobot) | | **[nanoclaw](https://github.com/qwibitai/nanoclaw)** | Containerized agent runtime that calls Atomic Chat as an MCP tool. | [Skill guide →](https://github.com/qwibitai/nanoclaw/blob/main/.claude/skills/add-atomic-chat-tool/SKILL.md) | | **[OpenClaude](https://github.com/Gitlawb/openclaude)** | Open-source coding-agent CLI for cloud and local models. Lists Atomic Chat as a supported provider. | [Providers list →](https://github.com/Gitlawb/openclaude#supported-providers) | | **[Kilo Code](https://kilo.ai/)** | Open-source AI coding agent for VS Code, JetBrains, and CLI. Ships with first-class Atomic Chat provider support and auto-discovery. | [Setup guide →](https://kilo.ai/docs/ai-providers/atomic-chat) | | **[Hermes Desktop](https://github.com/fathah/hermes-desktop)** | Native desktop companion for Hermes Agent. Includes an Atomic Chat local preset at `http://localhost:1337/v1`. | [Repo →](https://github.com/fathah/hermes-desktop) | | **[Hermes Workspace](https://github.com/outsourc-e/hermes-workspace)** | Local-first agent workspace built on Nous Research's Hermes. Uses Atomic Chat as its inference backend. | [Repo →](https://github.com/outsourc-e/hermes-workspace) | > Built something that runs on Atomic Chat? [Open a PR](https://github.com/AtomicBot-ai/Atomic-Chat/pulls) and we'll add it here. --- ### Build from Source #### Prerequisites - Node.js ≥ 20.0.0 - Yarn ≥ 4.5.3 - Make ≥ 3.81 - Rust (for Tauri) - (Apple Silicon) MetalToolchain `xcodebuild -downloadComponent MetalToolchain` #### Run with Make ```bash git clone https://github.com/AtomicBot-ai/Atomic-Chat cd Atomic-Chat make dev ``` This handles everything: installs dependencies, builds core components, and launches the app. **Available make targets:** - `make dev` — full development setup and launch - `make build` — production build - `make test` — run tests and linting - `make clean` — delete everything and start fresh #### Manual Commands ```bash yarn install yarn build:tauri:plugin:api yarn build:core yarn build:extensions yarn dev ``` --- ### System Requirements - **macOS**: 13.6+ (8GB RAM for 3B models, 16GB for 7B, 32GB for 13B) - **Windows**: 10/11 x64 (same RAM recommendations as macOS) - **Linux**: x86_64, glibc ≥ 2.35 (Ubuntu 22.04+, Debian 12+, Fedora 40+, Arch, Mint, Pop!_OS — same RAM recommendations as macOS). Optional: a Vulkan loader (`vulkan-1` package, or `mesa-vulkan-drivers` / proprietary NVIDIA driver) for GPU acceleration. - **iOS**: 17+ (download from App Store) - **Android**: download from Google Play --- ### Running on Linux Atomic Chat ships as a single self-contained `.AppImage` — no installer, no root: ```bash chmod +x Atomic.Chat_*_amd64.AppImage ./Atomic.Chat_*_amd64.AppImage ``` If prompted about FUSE on first launch: `sudo apt install fuse libfuse2` (Debian/Ubuntu) or `sudo dnf install fuse fuse-libs` (Fedora). GPU acceleration (Vulkan) is auto-detected on first launch; only GGUF models run on Linux. --- ### Troubleshooting If something isn't working: 1. Copy your error logs and system specs 2. Open an issue on [GitHub](https://github.com/AtomicBot-ai/Atomic-Chat/issues) 3. Or ask for help in our [Discord](https://discord.com/invite/8wGSsvmg4V) --- ### Star History <a href="https://star-history.com/#AtomicBot-ai/Atomic-Chat&Date"> <img src="https://api.star-history.com/svg?repos=AtomicBot-ai/Atomic-Chat&type=Date" width="100%" alt="Star History" /> </a> --- ### License Apache 2.0 — see [LICENSE](LICENSE) for details. ### Acknowledgements Built on the shoulders of giants: - [llama.cpp](https://github.com/ggml-org/llama.cpp) - [MLX-VLM](https://github.com/Blaizzy/mlx-vlm) - [Tauri](https://tauri.app/) - [Scalar](https://github.com/scalar/scalar) --- <p align="center"> <sub>© 2026 Atomic Chat · Built with ❤️ · <a href="https://atomic.chat">atomic.chat</a></sub> </p>