Free AI Coding Assistants

Free AI Coding Assistants (commercial terms & conditions may change dynamically): 

1. The Individual Tier of AWS CodeWhisperer is free for some IDEs.
2. TabNine's Starter plan offers basic code completion. Tabnine’s generative AI uses open-source code with permissive licenses for its Public Code trained AI model (MIT, Apache 2.0, BSD-2-Clause, BSD-3-Clause). Tabnine claims that whether you’re using Tabnine’s Pro plan or our Basic plan, your code, and AI data are NEVER used to train any models other than private code models.
3. MutableAI for VS Code can autocomplete, document & refactor your code. Its free tier for individuals and open source devs.
4. Blackbox VS Code Extension can autocomplete & search for code.
5. Safurai is available as a VS Code Extension. The current alpha version is free.
6. HF Code Autocomplete VS Code Extension from Hugging Face was forked from tabnine-vscode & modified for making it compatible with open source code models on hf.co/models.
7. IntelliCode - The basic version of IntelliCode is available as a free extension for Visual Studio Code and can be used with any programming language supported by Visual Studio or Visual Studio Code for AI-driven code auto-completion. IntelliCode Completions predicts up to a whole line of code based on your current context.

While the free options are useful to get started, commercial tools offer more powerful features.

AI Assistants can also help with understanding & writing code in popular languages outside the editor.

Some common types of machine learning models that are often used in AI code assistants:

• Language models which predict the next word or phrase in a sequence of text. 
• Deep neural networks suggest code completions based on the patterns and relationships in the code being written. Deep neural networks can be trained on large datasets of code to learn common coding patterns and generate code suggestions.
• Reinforcement learning models provide feedback on the quality of the code being written. Reinforcement learning models can learn to identify coding patterns that are likely to cause issues or errors, and provide recommendations for how to improve the code.
• Transformer models perform natural language processing (NLP) tasks such as language translation and text classification. Transformer models are trained on large datasets of text and can learn to generate accurate and natural-sounding translations and classifications.