Metadata-Version: 2.4
Name: lmms
Version: 2.1.11
Summary: Builder, hardware profiler, and toolchain manager for the LMMs Ecosystem.
Author: Raj Singh
Author-email: mr.rajsingh18110@gmail.com
Requires-Python: >=3.9
Description-Content-Type: text/markdown
Requires-Dist: psutil
Requires-Dist: rich
Requires-Dist: requests
Dynamic: author
Dynamic: author-email
Dynamic: description
Dynamic: description-content-type
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

# lmms

**Bootstrapper and Hardware Profiler for the LMMs Engine (Local Machine Model Studio)**

[![Documentation](https://img.shields.io/badge/Documentation-lmms.markanm.com-007A87?style=for-the-badge&logo=read-the-docs&logoColor=white)](https://lmms.markanm.com/)

`lmms` is the official setup, profiling, and installation utility for the LMMs ecosystem. It helps users prepare their systems for running local AI models by detecting hardware capabilities, performing compatibility checks, and installing the optimal LMMs Engine components.

## What It Does

Before running AI models locally, `lmms` analyzes your system and automatically configures the best environment for the LMMs Engine.

It is designed to:

- Detect system hardware and software capabilities.
- Evaluate compatibility for local AI inference.
- Download and install the appropriate LMMs Engine binaries.
- Configure dependencies and runtime environments.
- Diagnose common setup and performance issues.

---

## 📖 Documentation

For detailed guides, CLI command reference, configuration tutorials, and hardware requirements, please visit the official documentation website:

> [!NOTE]
> ### 🌐 Official Documentation
> You can view full documentation and ecosystem resources at:
> **[lmms.markanm.com](https://lmms.markanm.com/)**

---

## Features

### Hardware Profiling

Automatically detects:

- Operating System
- CPU Architecture
- Available RAM
- GPU Model(s)
- GPU VRAM
- CUDA Availability and Version
- Python Environment

### Compatibility Analysis

- Evaluates hardware readiness for local AI workloads.
- Provides compatibility information before installation.
- Helps identify potential bottlenecks.

### Intelligent Installation

- Downloads optimized engine binaries from GitHub Releases.
- Selects the most suitable runtime configuration based on detected hardware.
- Simplifies setup for both CPU-only and GPU-enabled systems.

### Diagnostics & Health Checks

Built-in tools for:

- Dependency verification
- Environment validation
- Configuration troubleshooting
- Performance diagnostics

### Global CLI Access

Installs the LMMs Engine and related tools into your system PATH, allowing commands to be executed from anywhere.

### Protected Distribution

The project uses Cython-based compilation and packaging techniques to help protect proprietary components.

---

## Installation

Install from PyPI:

```bash
pip install lmms
```

---

## Usage

### 1. Installer / Updater

**Interactive Architecture Wizard (Recommended)**
```bash
lmms install
```
Opens an interactive terminal flowchart wizard to choose your architecture setup:
- **Individual**:
  1. Engine 1 (`llama.cpp + AirEngine` - Fast & Low RAM, Port 11435)
  2. Engine 2 (`PyTorch + Transformers + AirEngine` - Full precision safetensors)
  3. Super Engine (`llama.cpp + PyTorch + AirEngine` - All 3 engines together)
  4. Desktop GUI Studio (folder: `gui/` + backend)
  5. Terminal CLI Studio (folder: `cli/` + backend)
  6. Complete Suite (All-in-One: GUI + CLI + Voice + Super Engine)
- **Enterprise** *(Features `[B / ESC]` back-navigation)*:
  - **Host Server Infrastructure**:
    1. Dedicated Super Host Server (All 3 engines serving team on LAN/VPC `0.0.0.0:11435`)
    2. PyTorch / AirEngine High-Performance Node (Multi-GPU & large model serving)
    3. Engine 1 Lightweight Node (`llama.cpp + AirEngine`)
  - **Remote Client Studio**:
    1. Enterprise Desktop Client (GUI with Local->LAN->Global auto-discovery)
    2. Enterprise Terminal Client (CLI with Local->LAN->Global auto-discovery)
    3. Complete Client Suite (Both Desktop GUI & Terminal CLI)

**System Check (Hardware Profiler & Prerequisites)**
```bash
lmms -check
```
Automatically detects your OS, hardware compatibility (CPU, GPU, RAM), installed LMMs components, and prerequisites.

**Install Entire Ecosystem (main branch)**
```bash
lmms install --all
```
Downloads and installs the complete LMMs ecosystem (GUI + CLI + Backend + Engine).

**Install GUI Only (gui branch)**
```bash
lmms install --gui
```
Downloads and installs the Desktop GUI Studio, Backend, and Engine.

**Install CLI Only (cli branch)**
```bash
lmms install --cli
```
Downloads and installs the Terminal CLI Studio, Backend, and Engine.

**Install Standalone Engine Only (engine branch)**
```bash
lmms install --engine
```
Installs the Super Standalone Engine (Llama.cpp + PyTorch + AirEngine) with pure API daemon on port 11435 (Zero GUI).
- `lmms install --engine1`: Standalone Engine 1 (Llama.cpp + AirEngine)
- `lmms install --engine2`: Engine 2 (PyTorch + AirEngine Server Node)
- `lmms install --host`: Dedicated Enterprise Super Host Server (0.0.0.0:11435)
- `lmms install --client-gui`: Enterprise Client GUI (with auto-discovery)
- `lmms install --client-cli`: Enterprise Client CLI (with auto-discovery)

**Smart Update**
```bash
lmms update
```
Smartly detects which component you installed and pulls latest updates from the corresponding branch.

**Update Specific Component**
```bash
lmms update --all | --cli | --gui | --engine
```

---

### 2. Uninstall

**Remove Source Code (Data Safe)**
```bash
lmms-uninstall -all
```
Deletes the LMMs source code but keeps your Python packages (VENV) and User Data (chats, models) safe.

**Total Purge (Factory Reset)**
```bash
lmms-uninstall -all --purge
```
Completely removes all LMMs code, Python environments, and user data from your PC.

---

### 3. Launch Options

**Run Specific Component**
```bash
lmms --gui
lmms --cli
lmms --engine
```
Directly runs the specified component.

**Run Default Configured Component**
```bash
lmms
```
Runs the default configured component.

---

## Workflow

```text
System Scan
     ↓
Hardware Profile
     ↓
Compatibility Analysis
     ↓
Binary Selection
     ↓
Engine Installation
     ↓
Diagnostics & Validation
```

---

## Project Vision

The goal of the LMMs ecosystem is to make local AI deployment simple, accessible, and hardware-aware.

Rather than requiring users to manually select CUDA versions, configure runtimes, and troubleshoot dependencies, `lmms` automates the entire setup process.

---

## Next Step

After setup is complete:

```bash
pip install lmms
```

or launch the installed LMMs Engine using the provided CLI tools.

---

## License

Developed by **MarkanM**.

See the LICENSE file for licensing information.

---

**Part of the LMMs Ecosystem — enabling seamless local, multi-model AI workflows across diverse hardware configurations.**
