IBM® watsonx™ Code Assistant

Overview

IBM® watsonx™ Code Assistant is an innovative, generative AI coding companion that offers robust, contextually aware assistance for popular programming languages including Go, C, C++, Java, JavaScript, Python, TypeScript, and more. Seamlessly integrated into your IDE, you can accelerate your productivity and simplify coding tasks, all with trust, security, and compliance.

Note for IBM Granite.Code users: IBM Granite.Code is now part of the watsonx Code Assistant product portfolio. For individual users, you can continue to use Ollama to access a local IBM Granite model. For increased performance and new features, provision a trial of watsonx Code Assistant on IBM Cloud. For more information, see the IBM Cloud catalog.

Features

Get code suggestions

Use chat conversations: Use natural language prompts to generate code suggestions. Use a chat conversation to enter a prompt that explains the code you need, and watsonx Code Assistant generates something you can choose to use.

Reference code: To ask questions or refine a specific file, class, function, or method in your workspace, you can use a code reference. These references provide important context and can help to increase the accuracy of the answer. As part of your chat message, type the @ symbol to see a list of files, classes, and methods from your workspace. Click to select the reference, and watsonx Code Assistant sends the contents of the reference as part of your message.

Code completion: Or, complete code in the editor. Start typing a line of code, then pause. IBM watsonx Code Assistant adds a code suggestion to complete the line that you typed.

You can also get a multiline code suggestion. Start typing a line of code,then use a keyboard shortcut, and watsonx Code Assistant adds a multiline code suggestion.

Or, enter a comment that describes the code you want.

For more information, see the documentation for Getting code suggestions.

Explain code

Use generative AI to analyze and summarize your code to understand what the code does. Click the Explain option that precedes a code block or enter /explain in a chat conversation. IBM watsonx Code Assistant analyzes the code and provides a detailed explanation of what the code does.

For more information, see the documentation for Explaining code.

Document code

Generate comment lines that document what your code does. Click the Document option that precedes a code block or enter /document in a chat conversation. IBM watsonx Code Assistant analyzes the code and adds comments that document what the code does.

For more information, see the documentation for Documenting code.

Generate unit tests

Create unit tests to evaluate your code functions. Click the Unit Test option that precedes a code block or enter /unit-test in a chat conversation. IBM watsonx Code Assistant analyzes the code and creates a unit test.

For more information, see the documentation for Generating unit tests.

Translate code from one language to another

Use watsonx Code Assistant to translate code. In a chat conversation, use the syntax translate [from <source language >] to <target_language> <code reference>

For more information, see the documentation for Translating code from one language to another.

Setup

Provision a watsonx Code Assistant service instance on IBM Cloud for your organization to get the best performance and the full set of features. Or for individuals, you can use Ollama to access a local IBM Granite model.

Use a watsonx Code Assistant service instance on IBM Cloud

To set up on IBM Cloud:

1. Use the IBM Cloud catalog to provision a service instance of watsonx Code Assistant.
2. When you finish provisioning your instance, click Setup to open an onboarding checklist page to help you with configuration.
3. Have your developers create an IBM Cloud API Key.
4. Have your developers install the Visual Studio Code extension for watsonx Code Assistant.

For more information, see the documentation for:

• Setting up your watsonx Code Assistant service in IBM Cloud
• Installing the IDE extension or plugin

Use watsonx Code Assistant with a local IBM Granite model

For individual users, IBM watsonx Code Assistant can access a local model through Ollama, which is a widely-used local inferencing engine for LLMs. Ollama wraps the underlying model-serving project llama.cpp.

Install the IBM watsonx Code Assistant extension

1. Open the IBM watsonx Code Assistant extension page in the Visual Studio Marketplace.
2. Click Install on the Marketplace page.
3. In Visual Studio Code, click Install on the extension.
4. In the extension settings, set Wca: Backend Provider to ollama.

Install Ollama

• MacOS, Linux, Windows: Download and run the ollama installer

• On MacOS, you can also use homebrew to install Ollama:

  brew install ollama
  

Start the Ollama inference server

In a console window, run:

ollama serve

Leave that window open while you use Ollama.

If you receive the message Error: listen tcp 127.0.0.1:11434: bind: address already in use, the Ollama server is already started.

Install the IBM Granite code model

Get started with IBM watsonx Code Assistant by installing the granite-code:8b model available in the Ollama library.

1. Open a new console window.

2. On the command line, type ollama run granite-code:8b to download and deploy the model. You see output similar to the following example:

   pulling manifest 
   pulling 8718ec280572... 100% ▕███████████████████████ 4.6 GB
   pulling e50df8490144... 100% ▕███████████████████████ ▏  123 B
   pulling 58d1e17ffe51... 100% ▕███████████████████████▏  11 KB
   pulling 9893bb2c2917... 100% ▕███████████████████████▏  108 B
   pulling 0e851433eda0... 100% ▕███████████████████████▏  485 B
   verifying sha256 digest 
   writing manifest 
   removing any unused layers 
   success 
   >>> 
   

3. Type /bye after the >>>to exit the Ollama command shell.

4. Try out the model by typing:

   ollama run granite-code:8b "How do I create a python class?"
   

5. You should see a response similar to:

   To create a Python class, you can define a new class using the "class" keyword followed by the name of the class and a colon. Inside the class definition, you can specify the methods and attributes that the class will have. Here is an example: ...
   

Configure the Ollama host

By default, the Ollama server runs on IP address 127.0.0.1, port 11434, andhttp as a protocol. If you change the IP address or the port where Ollama is available:

1. In Visual Studio Code, open the extension settings for watsonx Code Assistant.
2. In Wca > Local: API Host, add the host IP and port.

Configure the Granite model to use

By default, watsonx Code Assistant uses the granite-code:8b model for both chat and code completion. If your environment has enough capacity, install the granite-code:8b-base model.

To use a different model:

1. Install the granite-code:8b-base model. See Install the Granite code model.
2. In Visual Studio Code, open the extension settings for watsonx Code Assistant.
3. In Wca > Local: Code Gen Model, enter granite-code:8b-base.

Securing your setup

Your Visual Studio Code environment

IBM watsonx Code Assistant does not provide any additional security controls. It's recommended the following steps be taken to properly secure your setup:

• Apply all Visual Studio Code updates to help ensure you have the latest security and bug fixes. For more information, see the Microsoft Documentation.
• The IBM watsonx Code Assistant extension logs are stored in *.log files under <your home directory>/.wca. These files are not encrypted, other than the encryption that your file system provides. Safeguard the logs against improper access.

Chat conversation storage

IBM watsonx Code Assistant stores all your chat conversations locally in your file system under <your home directory>/.wca/chat.db, in a database format defined by SQLite. IBM watsonx Code Assistant does not share these conversations with anyone. This file is not encrypted, other than the encryption that your file system provides. Safeguard this file against improper access.

Telemetry data

IBM watsonx Code Assistant does not collect any telemetry data. In general, IBM watsonx Code Assistant does not send any data that it processes to a third party, IBM included.

Connecting IBM watsonx Code Assistant and Ollama

By default, the Ollama server runs on IP address 127.0.0.1, port 11434, using http as a protocol, on your local device. To use https instead, or go through a proxy server, see the Ollama documentation.