A native macOS AI chat app that runs entirely on your Mac. No cloud, no account, no subscription.
Klee uses MLX to run large language models directly on Apple Silicon, so your conversations never leave your device.
- 100% local inference -- your data stays on your Mac
- No account or API key required -- download and start chatting
- One-click model download -- pick a model, Klee handles the rest
- Streaming responses -- tokens appear as they're generated
- Lightweight -- native SwiftUI app, no Electron, no Docker, no background services
| Requirement | Minimum |
|---|---|
| macOS | 15.0 (Sequoia) or later |
| Chip | Apple Silicon (M1 or later) |
| RAM | 8 GB (see model table below) |
More RAM unlocks larger, more capable models:
| RAM | Recommended Models |
|---|---|
| 8 GB | Gemma 3 4B, Qwen3 4B, Phi 4 Mini |
| 16 GB | Qwen3 8B, Gemma 3 12B, DeepSeek R1 8B |
| 32 GB | Qwen3 14B, Qwen3 30B (MoE) |
| 64 GB+ | Gemma 3 27B, DeepSeek R1 32B |
Klee is distributed directly as a signed macOS app (Developer ID), not through the App Store.
- Download the latest
.dmgfrom Releases - Drag Klee into your Applications folder
- Open Klee -- if macOS shows a Gatekeeper warning, go to System Settings > Privacy & Security and click "Open Anyway"
- Open Klee -- the app detects your system RAM and shows compatible models
- Download a model -- tap the download button next to any recommended model. Downloads resume automatically if interrupted
- Start chatting -- select the downloaded model and type your message
Models are cached in ~/Library/Caches/models/ and persist across app restarts.
All models are 4-bit quantized variants from the mlx-community on HuggingFace.
| Model | Size | Min RAM | HuggingFace ID |
|---|---|---|---|
| Gemma 3 4B | ~3 GB | 8 GB | mlx-community/gemma-3-4b-it-qat-4bit |
| Qwen3 4B | ~2.5 GB | 8 GB | mlx-community/Qwen3-4B-4bit |
| Phi 4 Mini | ~2.2 GB | 8 GB | mlx-community/Phi-4-mini-instruct-4bit |
| Qwen3 8B | ~5 GB | 16 GB | mlx-community/Qwen3-8B-4bit |
| Gemma 3 12B | ~8 GB | 16 GB | mlx-community/gemma-3-12b-it-qat-4bit |
| DeepSeek R1 8B | ~4.6 GB | 16 GB | mlx-community/DeepSeek-R1-0528-Qwen3-8B-4bit |
| Qwen3 14B | ~8 GB | 32 GB | mlx-community/Qwen3-14B-4bit |
| Qwen3 30B (MoE) | ~17 GB | 32 GB | mlx-community/Qwen3-30B-A3B-4bit |
| Gemma 3 27B | ~17 GB | 64 GB | mlx-community/gemma-3-27b-it-qat-4bit |
| DeepSeek R1 32B | ~18 GB | 64 GB | mlx-community/DeepSeek-R1-Distill-Qwen-32B-4bit |
Requires Xcode 16+ and macOS 15.0+.
git clone https://github.com/signerlabs/Klee.git
cd Klee
open Klee.xcodeprojSelect the Klee scheme, then build and run (Cmd+R). SPM dependencies (mlx-swift-lm) will resolve automatically on first build.
MIT