DeepChat

v 1.0.5

DeepChat is a VS Code extension that integrates with the Ollama API and provides seamless AI-powered conversations using the DeepSeek-R1 model. This extension is designed to help developers and users interact with AI directly within the VS Code environment with minimal setup.

Features

• Interactive Chatbot UI: Provides an easy-to-use chatbot interface for conversational AI, similar to ChatGPT, directly within VS Code.
• Model Selection: Switch between different model sizes (1.5b, 8b, 32b) depending on your system's resources and needs.
• Automatic Model Management: The extension automatically downloads models if not already present, but requires initial setup for Ollama.
• Background Operation: The extension uses the VS Code command palette to manage and interact with the chatbot, while running Ollama in the background.

Requirements installation

Installing ollama

1. Go to the ollama download page and download ollama for your system.
2. Pull one of the three models using ollama pull deepseek-r1:<1.5b|8b|32b> Choose one of the following model sizes based on your system’s capabilities:

• 1.5b: Smallest and most resource-friendly.
• 8b: Balanced size for most systems.
• 32b: Largest and most powerful, requiring more system resources.

Note: The model download may take some time depending on your internet speed and the size of the model and can also take up some disk space.

Installing the extension

1. Press Cmd + P or Ctrl + P depending on your platform.
2. Type in or paste in ext install deepchat-r1 and press Enter. This will open the extension sidebar and you'll be able to just hit install.

Usage

Once you've finished installing all the requirements you can start chatting.

1. Open the command pallete using Cmd + Shift + P or Ctrl + Shift + P depending on your system.
2. Type in DeepChat: Set <1.5b/8b/32b> depending on what model you've pulled and hit enter.
3. Open the command pallete again and type in DeepChat: Open chat. This will open a sidebar with the chat.

You can also pull multiple models and play around with them. You can change them using DeepChat: Set <1.5b/8b/32b> in the command palette.

The selected model will be used for the next message you send.

How It Works

• DeepChat uses the Ollama API to handle the running of models locally on your machine.
• Once you’ve downloaded a model, all interactions are processed on your computer, providing a fast, private experience.
• The extension interacts with Ollama to download, run, and switch between models as needed.

Troubleshooting

• Model Download Issues: If the model isn’t downloading, ensure your internet connection is stable and try running the ollama pull command again.
• Slow Responses: If the AI responds slowly, consider switching to a smaller model like 1.5b if you’re running on a resource-constrained system.
• Ollama Not Running: If you encounter any issues, make sure that Ollama is running in the background.

Common Commands

• Open the Chat: Press Cmd + Shift + P (macOS) or Ctrl + Shift + P (Windows/Linux), then type deepchat: Open Chat and press Enter.
• Switch Models: Press Cmd + Shift + P (macOS) or Ctrl + Shift + P (Windows/Linux), then type deepchat: Set 1.5b/8b/32b and press Enter.

Future Features

• I am currently building on code autocompletion like with using Tabnine. It might still take a while but I am actively working on this project.
• I don't have many ideas so if you want to request a feature or report a bug, please go to the github page and do it there. Thanks!