GPT Everywhere: Desktop AI 2.2.9

GPT Everywhere allows you to easily leverage the power of AI through a user-friendly interface. With features like drag-and-drop file transfers, keyboard shortcuts, support for multiple models, and powerful output controls, this app aims to streamline your AI experience. Using your own API key, GPT Everywhere provides private GPT access tailored to your needs.

**Important: Requires your own OpenAI API key, Azure AI key, or OpenRouter.ai key.

GPT Everywhere Features

• Drag and drop functionality: Directly transfer project files into the GPT context.
• Shortcut: Launch GPT and generate feedback in any application.
• Supports many models: Easily convert from GPT-4 Turbo to Mythomax
• Custom specializations: Create different specializations or skill sets for GPT based on your requirements.
• Customizable output settings: Set maximum temperature and token parameters for AI-generated content.
• Whisper AI integration: Voice command your instructions to the app.
• Integrated Monaco editor: Develop, edit and manage your project files within the app.
• Task automation: Simplify complex workflows with threads for multi-step processes.
• Comprehensive User Guide: Detailed instructions to understand all the features and functions of the app.
• Using a secured API key: The app does not store your OpenAI API key.
• Encoding support: Directly create new files from code markup response.
• Testing localized AI models: Testing with models using Oobabooga.

What's new in this version

• February 2024: Dark mode is here! Google search using APIFY. Other minor improvements.
• January 2024: Gemini Pro supports the use of Vertex AI. Latest GPT-4 turbo model. Bug fixes and improvements. New Open Router Models
• December: Improved Azure support for DALL·E 3 and GPT-4 128k, Claude API support, and batch image processing with DALL·E 3. Vision model improvements. And new features like simple AI agents and easier system text generation.
• November: Introducing the GPT-4 128k model, creating images with DALL·E 3, experimenting with Text-to-Speech (TTS) and Visual Modeling.

System requirements

• OS: Windows 10 version 14316.0 or higher
• Architecture: x64, Intel only.