How to use Claude Code FREE Forever (Openrouter)
How to Use Claude Code for Free
Introduction to Claude Code
- The video introduces the concept of using Claude Code without incurring costs, highlighting its potential expense and limitations when credits run out.
- Presenter Jack Roberts shares his background in tech startups and AI automation, setting the stage for a practical tutorial on Claude Code.
Understanding Claude Code's Framework
- Claude Code is compared to an F1 car; switching models is akin to changing engines while retaining core benefits.
- Key trade-offs are discussed: cost, speed, performance, and security. Users can access some features for free but must understand these limitations.
Setting Up with Open Router
- To utilize Claude Code effectively, users need to set it up via Open Router, which provides access to various models.
- Five core models are introduced for selection based on user needs; free options may come with rate limits that could affect performance.
Model Options and Cost Analysis
- Various model costs are outlined: Sonic 4.6 ($15), Gemini Flash ($1.50), Dro small (cheaper), with two free options available.
- Recommendations are made based on project type; for client projects, using higher-quality models like Claude Opus or Sono is advised.
Performance Metrics of Models
- A comparison of five realistic model options includes their costs and performance ratings (e.g., Deep Seek V3 - free).
- Users can select models based on desired performance versus cost efficiency; Gemini Flash offers 85% quality at a fraction of the price.
Final Steps in Setup
- Instructions are provided for creating an API key in Open Router after selecting a model like Gemini Flash to maximize coding duration at lower costs.
How to Set Up and Use Claude with Anti-Gravity
Setting Up the Model
- The speaker discusses the importance of investing in a paid model, suggesting that spending $10 can significantly enhance the user experience.
- Instructions are provided for creating a new model called "JR London" with a credit limit set at $15, emphasizing ease of setup.
Using Anti-Gravity
- The speaker prefers using Anti-Gravity for its ability to maintain conversations with Gemini without interference from other models.
- A step-by-step framework is introduced for configuring the model, including copying an API key and setting up task lists.
Testing and Switching Models
- The configuration allows switching requests from Anthropics to Open Router easily, enhancing flexibility in model usage.
- After testing commands in the terminal, itโs confirmed that they are interacting with Googleโs Gemini 7.5 preview model.
Changing Models
- The process of changing models is demonstrated by copying prompts into Anti-Gravity to switch between different free modules seamlessly.
- Users are prompted whether they want to change models; upon confirmation, everything transfers smoothly.
Terminal Tips and Functionality
- Closing and reopening the terminal is necessary for changes to take effect; users are advised on efficient command cycling using keyboard shortcuts.
- The speaker highlights how Open Router Free cycles through various free models, addressing potential rate limits encountered when using specific ones.
Leveraging Skills within Models
- Upon reopening Anti-Gravity, users can access skills within different models like Gemini 3 flash preview for enhanced functionality.
How to Use Bitly Skills in Opus
Introduction to Bitly Skill
- The speaker introduces the Bitly skill, which converts a given URL into a shortened Bitly link.
- An example is provided using the URL "glido.com" to demonstrate how the skill functions.
Functionality and Portability
- The speaker mentions that if you run out of tokens, your skills become portable, allowing for flexibility in usage.
- After successfully creating the Bitly link, the speaker verifies its functionality by checking if it redirects correctly to Glidos.
Managing Different Models
- A new browser panel is recommended for managing different AI models (e.g., Claude, Gemini), facilitating coding tasks.
- Instructions are given on reverting back to Claude by copying code from a reset prompt and pasting it into Opus.
Editing Settings
- The speaker discusses editing settings at both project and global levels through a
settings.jsonfile.
- Demonstration of opening a new terminal shows how to confirm which model (Opus 4.6 or Claude) is currently active.
Investment in AI Tools
- The speaker recommends considering the Claude Max plan as an investment rather than just a subscription cost.
- Emphasizes that hiring AI tools can be more cost-effective compared to traditional employment costs.
Value of Paid vs Free Models
- While free models can be helpful, they may lead to time loss due to mistakes; thus, investing in premium models is advised for efficiency.
- Encourages viewers to assess their specific needs when choosing between free and paid options for coding assistance.