Skip to main content

⚡ InfiniThink

A local-first desktop AI agent — no cloud, no subscriptions, 100% private.

Python PySide6 Ollama License


📖 Overview

InfiniThink lets you control your computer using plain English. Type a message or speak a command, and the AI interprets it, plans the steps, and executes them — all on your machine.

You: "prepare my research workspace"
InfiniThink: ✅ Launched Chrome
             ✅ Launched Notion  
             ✅ Opened research folder

Everything runs locally through Ollama (e.g. Llama 3). No API keys, no internet required for AI inference.


🏗️ Architecture

User (GUI / Voice / CLI)
        │
        ▼
┌──────────────────────┐
│  Input Processor     │  — Chat widget, STT, CLI readline
└────────┬─────────────┘
         │
         ▼
┌──────────────────────┐
│ Command Interpreter  │  — LLM via Ollama REST API → JSON command
└────────┬─────────────┘
         │
         ▼
┌──────────────────────┐
│   Task Planner       │  — Multi-step plans for complex requests
└────────┬─────────────┘
         │
         ▼
┌──────────────────────┐
│  Execution Engine    │  — Tool registry dispatch + error handling
└────────┬─────────────┘
         │
         ▼
┌──────────────────────┐
│    Tool System       │  — app_tools · file_tools · system_tools
└──────────────────────┘
         │
         ▼
   System Actions

See ARCHITECTURE.md for a full layer-by-layer breakdown.


📂 Project Structure

.
├── infini_think/           ← Core package (gui, core, tools, voice)
├── assets/                 ← Application icons and media
├── tests/                  ← Unit testing suite
├── .gitignore              ← GitHub exclusion rules
├── requirements.txt        ← Standard runtime dependencies
├── README.md               ← Main documentation
├── LICENSE                 ← MIT License (Private/Internal)
├── ARCHITECTURE.md         ← Technical architecture breakdown
└── pyproject.toml          ← Python packaging and entry points

🚀 Installation

Prerequisites

Requirement Install / Notes
Python 3.10+ python.org
Ollama ollama.com
Llama 3 model ollama pull llama3

1 — Clone and install

git clone https://github.com/your-org/infini-think.git
cd infini-think
pip install -r requirements.txt

Windows PyAudio note: If pip install pyaudio fails, install the pre-built wheel:

pip install pipwin && pipwin install pyaudio

2 — Start Ollama

ollama serve           # start the server (new terminal)
ollama pull llama3     # download the model (first time only)

3 — Launch InfiniThink

# GUI (default)
python -m infini_think.app.launcher

# Or install as a package and use the CLI command
pip install -e .
infini-think

💬 Usage

GUI Mode (default)

Launch the chat window and type or speak your commands:

Command What happens
open chrome Launches Google Chrome
organize my downloads Sorts Downloads folder by type
open my project folder Opens user's project directory
prepare my research workspace Opens Chrome + Notion + research folder
create a folder called Work Creates ~/Work
what's my system info Returns CPU, RAM, OS details
run ipconfig Executes shell command and shows output

CLI Mode

infini-think --cli
# or
infini-think --cli --verbose   # show plan details

Voice Mode

infini-think --voice

Override Model

infini-think --model mistral
infini-think --model llama3.1

🛠️ Development Setup

# Install dev dependencies
pip install -r requirements-dev.txt

# Run tests
python -m pytest tests/ -v

# Type checking
mypy Project/

# Linting
ruff check Project/

# Format
black Project/

🔧 Configuration

All settings in config/settings.py can be overridden via environment variables:

Variable Default Description
INFINI_OLLAMA_URL http://localhost:11434 Ollama server URL
INFINI_OLLAMA_MODEL llama3 Model to use
INFINI_OLLAMA_TIMEOUT 60 Request timeout (seconds)
INFINI_VOICE true Enable voice I/O
INFINI_THEME dark UI theme (dark/light)
INFINI_LOG_LEVEL INFO Log level

🔭 Roadmap

  • Plugin system — drop-in tools via Python entry points
  • Contextual memory — conversation history across sessions
  • Background agents — scheduled and event-triggered automation
  • Premium UI & Themes — dynamic Dark/Light mode toggle
  • Custom hotkey — global shortcut to open InfiniThink
  • Windows tray icon — always-on background mode

📄 License

MIT — see LICENSE.

Metadata

Release files for infini-think 1.0.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for infini-think 1.0.0
File Size Uploaded Provenance
infini_think-1.0.0.tar.gz 71.3 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for infini-think 1.0.0
File Interpreter ABI Platform
infini_think-1.0.0-py3-none-any.whl Python 3 none any Details

Total release size: 153.4 kB

Release files / infini_think-1.0.0.tar.gz

Download URL infini_think-1.0.0.tar.gz
Size 71.3 kB
Tags Source
SHA-256 checksum
How to use checksums
bb16f355cdb05074736f314b57201b07f19d3a7e9af0eacc705b0104e3959b18
BLAKE2b-256 checksum
How to use checksums
1fed40b6904678f0d00296b63644518e6392fef939e72e2d45a9c823b375dcc0
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.11.9

Release files / infini_think-1.0.0-py3-none-any.whl

Download URL infini_think-1.0.0-py3-none-any.whl
Size 82.2 kB
Tags Python 3
SHA-256 checksum
How to use checksums
889a2ff6c0a83b45fbe99e55206b8fff91b5878a3a67eaef33ca5a456c3567aa
BLAKE2b-256 checksum
How to use checksums
7bf7ba7ea6f2911e1aca2d4cd2b3bc24916ce920720c9f43e99506b6a1f53e9a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.11.9

Security for infini-think 1.0.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

1.0.0 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page