Trae AI 2.0: FULLY FREE AI Coder IS AMAZING! AGENTS, MCP, & UNLIMTED API Usage!

Key Concepts

Agentic IDE, Trey AI, Gemini 2.5 Pro, Cloud 3.7 Sonnet, AI agents, MCP (Model Context Protocol), Langatch, VS Code plugin, JetBrains plugin, autonomous file editing, browser automation, web scraping, code generation, full-stack development, AI-human collaboration.

Trey AI: An Agentic Code Editor

• Introduction: Trey AI is presented as a powerful, free, agentic code editor that has undergone significant upgrades. It facilitates AI-human collaboration in software development.
• Cost and Data Usage: While Trey AI offers free access to advanced models like Gemini 2.5 Pro and Cloud 3.7 Sonnet, it uses user data for training purposes. This might be a concern for proprietary or sensitive information but is less of an issue for personal projects or learning.
• Core Functionality: Trey AI functions as both a code editor and an autonomous AI agent system. It enables users to build, debug, and complete projects faster.
• Agentic Capabilities: The IDE has built-in agentic capabilities, allowing it to autonomously edit and delete files, and execute terminal commands. Users can configure and customize their own AI agents.
• Context and Accuracy: Trey AI can understand the context of the entire codebase, enabling it to make changes across multiple files.

New Features and Updates

• Plugins: Trey AI now has plugins for VS Code and JetBrains.
• Agent Rules and Tools: New agent rules and tools are available for defining agents.
• MCP Integration: Integration with MCP (Model Context Protocol) enhances agent functionality.

Langatch Sponsorship

• Problem: The video highlights the challenges of manually tweaking prompts and running test cases for AI models.
• Solution: Langatch is introduced as an open-source LLMOps platform designed to help AI teams deploy faster and smarter.
• Features:
  • Optimization Studio: Automates the process of finding the best prompt or model.
  • Evaluation: Evaluates LLM outputs, predicts effective paths, and allows non-technical experts to collaborate.
  • Observability: Provides full observability, batch testing, and cost tracking for developers.
  • Metrics Tracking: Tracks latency, cost, performance, and compliance metrics.
  • Security: Offers self-hosting or Langatch cloud with enterprise-grade security (GDPR and ISO compliant).
• Benefits: Langatch enables faster deployment (8x), optimization, and production-grade LLM pipelines.

Installation and Setup

• Installation: Trey AI can be installed with a one-click installer for Mac or Windows (Linux support is not currently available).
• Initial Setup: After installation, users are guided through an introduction and then directed to the main dashboard.
• Extension Store: Users can integrate extensions like Rootcode Client, Docker, and Remote Explorer.
• Codebase Upload: Users can upload their codebase to the IDE.

Agent Interaction and Model Selection

• Agent Interaction: The right-hand side of the IDE allows users to interact with the agentic Trey AI agent.
• Agent Customization: Users can create and customize their own AI agents with specific tools.
• MCP Integration: Agents can be built with MCP to add additional functionality.
• Model Selection: Users have free access to various models, including Gemini 2.5 Pro.
• Example Task: The video demonstrates generating a SaaS landing page using Gemini 2.5 Pro. The generated code can be applied to a new file or inserted into an existing one.

Deep Dive into Features

• Agent Builder: The agent builder allows users to create agents using natural language.
• Builder with MCP: This feature allows users to add MCP servers to enhance agent capabilities.
• MCP Explanation: MCP (Model Context Protocol) is an open standard that enables AI agents to securely connect with external servers like Fetch or Gmail, enhancing their functionality and contextual understanding.
• Puppeteer MCP Example: The video demonstrates using the Puppeteer MCP for browser automation and web scraping. The agent scrapes content from the Trey AI website and outputs it as markdown.
• Autonomous Execution: Commands can be set to execute autonomously, but the presenter recommends maintaining oversight.
• Additional Features: The IDE supports web search capabilities, referencing docs, and workspaces.

Full-Stack Application Generation

• Example: The video showcases the creation of a full-stack task management application.
• Workflow: Edits can be reviewed and accepted individually or all at once.
• Features: Autocomplete, inline editing, and terminal command execution are supported.
• Result: A functional task management application with login, front-end, and task management capabilities is generated.

Conclusion

• Recommendation: Trey AI is recommended for prototyping personal projects due to its free access to state-of-the-art models and powerful agentic capabilities.
• Call to Action: Viewers are encouraged to explore Trey AI and other resources linked in the description.