How to Make Your Development Experience Smoother with llm-ls

September 12, 2020

Welcome, fellow developers! If you’re on a quest to enhance your coding environment with the help of Large Language Models (LLMs), you’re in the right place. Today, we’re diving into how you can leverage **llm-ls** to create a more efficient software development experience.

What is llm-ls?

llm-ls is a Language Server Protocol (LSP) server that leverages the power of LLMs to significantly improve your development workflow. The goal here is to provide an accessible platform for IDE extensions, offloading the heavy lifting to **llm-ls** so that your extension code remains lightweight and manageable. This means you can focus on coding rather than on configuration.

Features of llm-ls

Prompt Generation: Uses the current file’s content to generate prompts, ensuring they are within the model’s context window by tokenizing the input. You can also choose whether to fill in the middle of your context or not!
Telemetry: Gathers data on requests and completions to assist in retraining the model while ensuring that your information is safe and stored locally in your system (~.cache/llm_ls/llm-ls.log).
Completion: Parses the Abstract Syntax Tree (AST) of the code to determine if it should return single-line, multi-line, or no completions at all.
Multiple Backends: Compatible with various APIs like Hugging Face’s Inference API, Ollama, and OpenAI-compatible APIs, allowing flexibility in your development process.

Getting Started with llm-ls

To integrate llm-ls into your IDE, follow these simple steps:

Ensure you have the required dependencies installed. This includes the LSP client for your chosen IDE (e.g., llm.nvim for Neovim, llm-vscode for Visual Studio Code, and llm-intellij for IntelliJ).
Set up the llm-ls server, ensuring it has access to the necessary LLM models.
Configure your IDE to connect with the llm-ls server and begin coding!

Troubleshooting Common Issues

Like all software, you may run into some bumps along the way. Here are a few common issues and their solutions to keep your development smooth:

Issue: The prompts generated aren’t relevant to my current file context.
Solution: Ensure that the prompt generation settings are properly configured within your IDE, allowing llm-ls to use the correct context.
Issue: Data appears to be missing from the log.
Solution: Double-check the logging settings and ensure the log level is set to info.
Issue: Incompatibility with certain APIs.
Solution: Ensure that the backend service you’re attempting to connect to is supported and properly configured.

For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.

Conclusion

At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.