8 must-have LM Studio plugins to level up your local LLM

Photo: Jacob Bobo

Large language models (LLMs) that can be hosted on local hardware are improving just as quickly as frontier models that can only be accessed online.

The Text Arena leaderboard currently ranks Google's Gemma-4-26-A4B just behind OpenAI's GPT 5.2 and just above GPT 5.1. Gemma-4-31B ranks even better—it's just behind Claude Sonnet 4.5 and beats OpenAI's GPT 5.3.

But it's not as simple as that. If you download LM Studio and install the best LLM your PC can handle, you're likely to be disappointed by the experience. That's because the model doesn't have access to many tools by default. It can't search the web, run shell commands, or write and edit files, among other things.

That's where plugins come in. A relatively new feature to LM Studio (in fact, it's still in beta), plugins provide an easy way to add tools for your local LLM. Installing plugins will upgrade LM Studio to include most of the functionality found in ChatGPT and Claude.

Some things you should know before installing plugins

Plugins are a potential security risk. That's not at all exclusive to LM Studio, but there's a particular trade-off that local AI tools must deal with. The broader the permissions you provide, the more useful the tool can be.

Some of these plugins have the ability to write to files or execute shell commands. LM Studio plugins will ask permission before they call tools, and while it might be tempting to select the auto-approve option, I recommend that you require approval for any tasks that involve modifying files or executing commands. Giving plugins complete approval to execute commands without permission will make your PC vulnerable to prompt injection attacks.

Fortunately, plugins can be turned on or off at any time. It's a good idea to only enable the plugins required to accomplish a task. Having too many tools enabled can cause local LLMs to fall into loops as they become confused by the number of tools at their disposal.

Beledarian's LM Studio Tools

While this is technically just one LM Studio plugin, it actually includes the functionality of a couple dozen. This plugin includes web search, browser navigation, sub-agents, memories, command line tools, file access, and JavaScript and Python code execution, among others.

It's something of a one-stop-shop, and if you download just one plugin, this is the one to get. You'll find your local LLM can access online content with ease, browse local files, and even make code changes into files or write new files (though remember to be cautious about allowing full automation of file changes and shell commands).

The result is that LM Studio gains the ability to do far more complex things than it can out of the box. You can, for example, vibe code a program from start to completion, generating code directly to files on disk that you can execute on your local machine.

DuckDuckGo and Visit-Website

Beledarian's LM Studio Tools is a useful plugin, but it's also a lot—perhaps more than you need or want. If you'd like to stick to the basics, and build from there, then I suggest you start with the DuckDuckGo and Visit-website plugins from Danielsig.

The DuckDuckGo plugin (which, fun fact, appears to be the most popular LM Studio plugin to date) provides access to the titular search engine. This is even more useful than it sounds. LLMs can parse data much more quickly than any human, so this tool can ingest a dozen results as quickly as you can type a single query in the search field.

However, the DuckDuckGo plugin is only a search tool. It can't access websites—which is where the Visit-website plugin comes in. When these plugins combine, they become a powerful web search tool.

Reddit Search

Continuing the web search theme is Reddit Search, which does exactly what it says on the tin. This plugin can search Reddit and browse Reddit posts to find relevant information. The quality of advice on Reddit isn't exactly the gold standard for...well, anything. However, it's often the best source for leaping into research on a niche topic. This tool lets the LLM do that for you.

Save to Excel

Some local LLMs really, really like to generate tables (I'm looking at you, GPT-OSS…). However, copying those tables to another document can be a chore. The Save to Excel plugin helps with this issue by turning tables into .XLSX files you can open in Excel or other compatible spreadsheet software. This can be a great way to extract data from a conversation even if you don't need an Excel file, because Excel can export many different formats.

Todo list

Unlike some of the prior plugins, this one isn't what it seems. You might expect this to be a tool for generating to-do lists based on a conversation, but that's not right. Instead, it creates a to-do list that a model can use to manage and track tasks. Think of it like a bargain bin version of the "planning mode" found in Claude. The creator of this plugin didn't stick with the project, but it's still useful, at least until someone decides to fork and maintain a version of it.

Computer

The Computer plugin is, shall we say, ambitious. It provides your local LLM with access to a Linux CLI to get down with however it wants. The Linux install runs in a container, providing a nice sandboxed environment—which is quite useful if you want the LLM to write to or edit files frequently. This plugin requires a bit more work than most to get started because you also need a container, like Docker, configured. However, it's worth the effort if you want to give an LLM broad access to a Linux command line.

Big-Rag

Retrieval augmented generation is a tool that allows an LLM to generate a response with context from your documents. LM Studio has a default RAG plugin, but Big-Rag adds broader file support and can address many large files. This is a must-have if you want to use a local LLM with a treasure trove of PDFs or other weighty documents.

Revision

LM Studio plugins are in beta though, to be honest, it's more like an alpha. One of the features currently missing is a way to update plugins you already have installed. Revision can't quite do that, but it can scan the plugins you have installed and then notify you if any updates are found. You still have to manually download the plugin, however.

Conclusion

These are far from the only plugins that are available, but they're the ones I recommend, as I've had a chance to use each of them and found them to be functional and useful. As mentioned, however, LM Studio's plugins are in beta, so keep an eye out for new plugins and features in the near future.

More from MC News

• Run AI Locally: The Best LLMs for 8GB, 16GB, 32GB Memory and Beyond
• Quantization Explained: Why the Same LLM Gives Better Results on High-End Hardware
• Why Context Length, Not AI Model Size, Is Your Biggest VRAM Problem
• How to Use LM Studio to turn your PC into an AI Server
• Why VRAM and Memory Bandwidth are Key for Powering Local AI
• Keyboard 101: Intro to Computer Keyboards
• Fix It Yourself: Talking to iFixit on Why Repairable Tech Matters
• The Local LLM Your Laptop's Been Waiting For
• Gemma vs Qwen: Which Local AI is Best?

Matthew S. Smith is a prolific tech journalist, critic, product reviewer, and influencer from Portland, Oregon. Over 16 years covering tech he has reviewed thousands of PC laptops, desktops, monitors, and other consumer gadgets. Matthew also hosts Computer Gaming Yesterday, a YouTube channel dedicated to retro PC gaming, and covers the latest artificial intelligence research for IEEE Spectrum.