# Unleash the Power of Choice: Using Third-Party LLMs in JetBrains AI Assistant with ProxyAsLocalModel

## Unleash the Power of Choice: Using Third-Party LLMs in JetBrains AI Assistant with ProxyAsLocalModel

The JetBrains AI Assistant is a powerful tool that integrates directly into your IDE, offering features like code completion, refactoring suggestions, and code explanation, all powered by large language models (LLMs). However, users might be limited to the default LLM provided by JetBrains. A new project, “ProxyAsLocalModel,” is changing that, allowing developers to leverage third-party LLM APIs within their familiar JetBrains environment.

The project, highlighted in a recent Show HN post, provides a clever workaround. Instead of relying solely on the built-in LLM options, ProxyAsLocalModel acts as a bridge, tricking the JetBrains AI Assistant into thinking a third-party LLM API is a locally running model. This opens a world of possibilities for developers who want to experiment with different LLMs, fine-tune their performance, or simply prefer a specific API for cost or feature reasons.

The technical details, found on the project’s GitHub repository (https://github.com/Stream29/ProxyAsLocalModel), likely involve setting up a proxy server that intercepts requests from the JetBrains AI Assistant and forwards them to the chosen third-party LLM API. The proxy then translates the LLM’s response into a format compatible with the JetBrains AI Assistant, creating a seamless integration for the user.

The advantages of this approach are numerous. Developers can now:

* **Explore a wider range of LLMs:** Experiment with different models like those offered by OpenAI, Anthropic, Google, or others, and find the perfect fit for their specific needs.
* **Optimize for cost:** Compare pricing models and select an LLM API that offers the best value for their usage.
* **Customize and fine-tune:** Some LLM APIs allow for fine-tuning on custom datasets, enabling developers to tailor the AI Assistant’s performance to their specific project or coding style.
* **Address privacy concerns:** Depending on the third-party API’s data handling policies, users might have more control over their data compared to relying solely on the default LLM.

While the initial setup might require some technical expertise, the potential benefits for developers are substantial. ProxyAsLocalModel represents a significant step toward more flexible and customizable AI-powered development environments. It empowers users to break free from vendor lock-in and embrace the full potential of the rapidly evolving LLM landscape directly within their JetBrains IDE. This project is definitely one to watch for developers seeking greater control and customization over their AI-assisted coding experience.