https://github.com/smartloop-ai/smartloop
On-device personal AI assistant
https://github.com/smartloop-ai/smartloop
ai auto-tuning fine-tune-llms fine-tuning framework gemma gguf llm llm-inference textual tui tui-app
Last synced: 2 days ago
JSON representation
On-device personal AI assistant
- Host: GitHub
- URL: https://github.com/smartloop-ai/smartloop
- Owner: smartloop-ai
- License: gpl-3.0
- Created: 2024-06-02T08:39:37.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2026-03-26T16:09:01.000Z (8 days ago)
- Last Synced: 2026-03-27T06:42:05.028Z (7 days ago)
- Topics: ai, auto-tuning, fine-tune-llms, fine-tuning, framework, gemma, gguf, llm, llm-inference, textual, tui, tui-app
- Language: Python
- Homepage: https://smartloop.ai
- Size: 348 KB
- Stars: 3
- Watchers: 1
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
An AI assistant to generate information and auto-tune from documents
[](https://smartloop.ai)
---
### Installation
Copy and paste the following script to your terminal to get started:
```bash
curl -fsSL https://smartloop.ai/install | bash
```
Install using [Homebrew](https://brew.sh) (recommended):
**macOS**
```bash
brew tap smartloop-ai/smartloop
brew install smartloop
```
**Linux**
```bash
# Install Homebrew for Linux first (if not already installed)
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
brew tap smartloop-ai/smartloop
brew install smartloop
```
> [!TIP]
> To upgrade: `brew update && brew upgrade smartloop`
**From source:**
> [!NOTE]
> Requires Python 3.11. For NVIDIA GPU acceleration, install [CUDA 12.6](https://developer.nvidia.com/cuda-12-6-0-download-archive) before proceeding.
```bash
git clone https://github.com/smartloop-ai/smartloop.git
cd smartloop
pip install -r requirements.txt
python main.py run
```
### Uninstall
```bash
# If installed via Homebrew
brew uninstall smartloop
brew untap smartloop-ai/smartloop
# If installed via curl
curl -fsSL https://smartloop.ai/uninstall | bash
```
### Usage
```bash
# View available commands
slp --help
# Initialize a new project
slp init -t
# Add a document
slp add document.pdf
# Run interactive chat
slp run
# no tui
slp run --no-tui
```
### Project Management
```bash
slp projects create
slp projects list
slp projects switch
slp status
```
### Server Management
SLP includes a background API server compatible with OpenAI's chat completion format.
```bash
slp server start
slp server stop
slp server status
```
On macOS, the server can also be managed via `brew services` (if installed using Homebrew):
```bash
brew services start smartloop
brew services stop smartloop
```
On Linux/WSL, the installer creates a systemd user service:
```bash
systemctl --user start smartloop
systemctl --user stop smartloop
systemctl --user status smartloop
```
### Requirements
| Requirement | Description | Required |
|-------------|-------------|----------|
| OS | macOS (Apple Silicon) or Linux (x86_64) or WSL | Yes |
| Python | 3.11+ | Yes |
| CUDA | 12.6+ (NVIDIA GPU acceleration) Metal | No |
| Metal| Bespoke on mac | yes |
### Troubleshooting
#### GPU not detected / Falls back to CPU
If the app falls back to CPU on a GPU-enabled system:
1. **Enable persistence mode:**
```bash
sudo nvidia-smi -pm ENABLED
```
2. **Verify GPU detection:**
```bash
nvidia-smi
```
If issues persist, ensure NVIDIA drivers are properly installed.
### License
© 2016 Smartloop Inc.
All code is licensed under the GPL, v3 or later. See [LICENSE](LICENSE) file for details.