Kite VS. TabNine: Which AI Code Autocomplete Should You Choose?
I tested two of the hottest AI code assistant in 2021
Imagine that when you use a development tool to program, you only need to enter one or two letters, and the tool will auto-generate the following code for you. It sounds like a kind of magic. In recent years, with the development of deep learning, there is no longer a kind of science fiction, and some development tools have already the prototype of such.
From 2019 to 2021, two popular AI Code Autocomplete, TabNine and Kite, came to my attention. After a period of use, I decided to do a brief review of them to evaluate which one is more suitable for me. I also hope that this article can help you when you choose an AI Code Assistant.
Preparation
First of all, this test is based entirely on my personal daily coding scenario: Using Python in VS Code, working on Machine Learning and Deep Learning related development. So this will be a subjective test.
The two of the most high-profile AI Code Autocomplete plugins I’m going to test are TabNine and Kite.
TabNine was released in 2019 that is based on GPT-2, which uses the Transformer network architecture and trained on 2 million source code from GitHub.
