I built my own Free VScode Copilot!!!

---

---

AI Summary

Summary of Video Transcript

• The video demonstrates how to use a local AI copilot for coding in Visual Studio Code without a subscription, using the AMA (AI Model Access) and Deep Seek Coder model.
• Instructions are provided for installing AMA, downloading the Deep Seek Coder model, and testing it with a simple Python addition code.
• The video guides through installing the Visual Studio Code extension ‘Continue’, which allows the use of various language models (LLMs).
• Detailed steps are given for configuring Continue to use the Deep Seek Coder model by editing the config.js file.
• The video shows how to use keyboard shortcuts within Visual Studio Code to interact with the Continue extension and how to switch between different models.
• Tips are provided on how to download and use other models like Stable LM Zero, and how to ensure they are accessible locally through ol serve.
• The video also covers privacy settings related to telemetry data sharing with Continue.dev.
• The process of writing, editing, and running Python code using the AI models is demonstrated, highlighting the differences in output quality based on the model used.
• The tutorial concludes with an emphasis on the benefits of using local models for sensitive data and the flexibility of using different models.

Detailed Instructions and Tips

• Install AMA.
• Download the Deep Seek Coder model with the command: AMA run deep-seek-coder.
• Install the Visual Studio Code extension ‘Continue’ from the marketplace or by searching for it in the Extensions tab.
• Configure config.js to use the Deep Seek Coder model by adding a block with the title, provider (AMA), and model name.
• Use keyboard shortcuts in Visual Studio Code with Continue:
  • Control or Command + M to select code.
  • Control + Shift + M for follow-up code selection.
  • Control + Shift + L for quick edit.
  • Control + Shift + R to debug the terminal.
• Download other models like Stable LM Zero with the command: AMA run stable-lm-zero.
• Ensure local access to models with ol serve.
• Disable telemetry data sharing in Continue’s extension settings if desired.
• Edit and run code using the AI models, and switch between them as needed.

URLs and Commands

• No specific URLs or CLI commands are provided in the summary.