YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
license: apache-2.0 language:
- fa
- en tags:
- cli
- terminal
- java
- custom-engine
- text-generation pipeline_tag: text-generation library_name: droidx
DroidX Engine 1.5
DroidX Engine 1.5 is a lightweight, standalone, and highly optimized Command-Line Interface (CLI) AI engine developed by CodeBloom. It is designed to provide a secure, offline, and highly responsive conversational AI experience directly within the terminal.
Built with a focus on performance and portability, the DroidX Engine encapsulates the entire model architecture, runtime, and tokenizer into a single, self-contained executable JAR file.
✨ Features
· 100% Offline & Private: No internet connection is required. All processing happens locally on your machine. · Self-Extracting Architecture: The engine automatically extracts its core model files to a hidden system directory (.droidx_system) upon the first run, keeping your workspace clean. · Stateful Conversations: Maintains chat history using unique session IDs, allowing for contextual and continuous dialogue. · Rich Terminal UI: Built-in support for ANSI colors and command history using the JLine library. · Optimized for Persian & English: Specifically tuned for high-quality Persian conversational output.
🛠️ System Requirements
To run the DroidX Engine, ensure you have the following installed:
· Java Development Kit (JDK) 21 or higher. · A terminal environment (Windows CMD/PowerShell, Linux, or macOS).
🚀 Execution Guide
- Download the app-all.jar file from this repository.
- Open your terminal and navigate to the directory containing the JAR file.
- Run the following command to start the engine:
java --add-modules jdk.incubator.vector --enable-preview -jar DroidX-Engine.jar
Terminal Commands
Once the DroidX-1.5 > prompt appears, you can interact using the following commands:
· : Type your message and press Enter to chat. · /new: Clears the current context and starts a brand-new chat session. · /exit: Safely shuts down the AI engine.
⚙️ Technical Specifications
· Engine Version: 1.5 · Runtime: Java 21 (Panama Vector API enabled) · Core Engine: Custom Jlama-based inference · Architecture: Encapsulated Model Context (EMC)
⚠️ Disclaimer
This engine is optimized for desktop and terminal environments. While it can run on mobile environments (e.g., Termux) via Java, performance may vary based on the device's hardware and available RAM.
Developed with precision by CodeBloom.