Among the many advances in artificial intelligence, code generation stands out as particularly notable. Coding patterns represent a major evolution in the field of software development.

Mistral AI has led the way in this transformation with its coding models, including the new Codestral 25.01designed to dramatically improve software development.

Table of contents

• Codestral 25.01
• Main features
• Obtain a Mistral AI API key
• Installing and configuring the Continue.dev extension in VS Code
• Using the Continue.Dev extension in VS Code

Codestral 25.01

Codestral 25.01 is the latest coding model released by mistral AI, representing a significant leap forward in AI-based software development.

Released on January 13, 2025, Codestral 25.01 had set a new benchmark in the field of code generation with its efficient and faster code generation capabilities.

Equipped with numerous features as well as a more efficient architecture and an improved tokenizer, it has now become the undisputed leader among its competitors.

Main features of Codestral 25.01

• It has a more efficient architecture and an improved tokenizer.
• It has been optimized for low latency and high frequency use cases.
• he masters more than 80 programming languages.
• It generates a response 2x faster than previous codestral models.
• It supports tasks such as fill-in-the-middle (FIM), code correction, and test generation.
• It can be easily integrated with popular IDEs.

Obtain a Mistral AI API key

1. Go to the Mistral AI website and create an account. You can either use your Google or Microsoft account or use your email address to register.
2. Log in to your account with your credentials and subscribe to a plan according to your needs.
3. Create an API key for your use, save the key for future use.

Installing and configuring the Continue.dev extension in VS Code

• Open the Extensions Marketplace in VS Code from the activity bar, search for “Continue AI” and install the extension.

• After installation, you can see a Continue button in the status bar at the bottom, click the button to open a pop-up window.

• Click Configure AutoComplete Options and open the config.json file.
• In the Config.json file, find the key “Autocomplete template tab” and you will find the following configuration.

"tabAutocompleteModel": [

    {
      "title": "codestral",
      "provider": "mistral",
      "model": "codestral-latest",
      "apiKey": ""
    }
  ]

• Paste your generated API key here between double quotes as shown below in the image.

"tabAutocompleteModel": [

    {
      "title": "codestral",
      "provider": "mistral",
      "model": "codestral-latest",
      "apiKey": "gfhs..........sfhus"
    }
  ]

Using the Continue.Dev extension in VS Code

1) Auto-complete code

• Open any file or a new file and start coding. The Continue.dev extension will provide autocomplete suggestions, as shown in the image.

• You can accept the suggestion by pressing the tab button on the keyboard or continue writing code accordingly.

2) Chat functionality

• On the activity bar on the left side of your window, you will see a new icon of the Continue.dev extension, click on it.
• Now you need to click on add chat template here.
• Now select Mistral from the Supplier drop-down list and Codestral from the Model drop-down list.

• Paste your generated API key into the API Key column and click connect.
• Once the template is successfully added, post your query in the chat and wait for the response.

That’s it! In this article, we aimed to explain how to configure the Continue.dev extension in VS Code using Codestral 25.01. Hope this helps! Please share your thoughts in the comments.

Start your machine learning journey with Webkul.
Happy coding!!

Tushar Sharma
4 badges

Passionate about machine learning, specialized in developing intelligent solutions using Python. I created this blog to share my journey, my projects and my knowledge about the world of machine learning. Join me to explore the exciting frontiers of AI and data science!