Upskill your LLMs With Gradio MCP Servers

Upskill your LLMs With Gradio MCP Servers
Have you ever ever wanted your favorite Large Language Model (LLM) to do greater than just answer questions? What if it could edit images for you, browse the net, or organize your email inbox?

Well, now it could actually! On this blog post, I’ll show you:

• What the MCP protocol is and the way it really works similarly to the smartphone apps we’re all used to, but for LLMs.
• How you will discover hundreds of MCP servers via the “MCP App Store.”
• How you can add one among these servers to your favorite LLM of selection to grant it a brand new ability. We’ll work through an example using Flux.1 Kontext[dev] which edits images from plain text instructions.

A Transient Intro To MCP

The Model Context Protocol (MCP) is an open standard that permits developers to construct secure, two-way connections between an LLM and a set of tools. For instance, in the event you create an MCP server that exposes a tool able to transcribing a video, then you definately can connect an LLM client (equivalent to Cursor, Claude Code, or Cline) to the server. The LLM will then know how you can transcribe videos and use this tool for you depending in your request.

Briefly, an MCP server is a normal solution to upskill your LLM by granting it a brand new ability. Consider it just like the apps in your smartphone. By itself, your smartphone cannot edit images, but you may download an app to do that from the app store. Now, if only there have been an app store for MCP servers? 🤔

Hugging Face Spaces: The MCP App Store

Hugging Face Spaces is the world’s largest collection of AI applications. Most of those spaces perform a specialized task with an AI model. For instance:

These spaces are implemented with Gradio, an open source python package for creating AI-powered web servers. As of version 5.28.0, Gradio apps support the MCP protocol.

That signifies that Hugging Face Spaces is the one place where you will discover hundreds of AI-powered abilities in your LLM, aka the MCP App Store!

Need to browse the app store? Visit this link. Manually, you may filter for MCP Compatible in https://hf.co/spaces.

An Example: An LLM that may edit images

Flux.1 Kontext[dev] is a powerful model that may edit a picture from a plain text prompt. For instance, in the event you ask it to “dye my hair blue” and upload a photograph of yourself, the model will return the photo but with you having blue hair!

Let’s plug this model as an MCP server into an LLM and have it edit images for us. Follow these steps:

1. Go to Hugging Face and create a free account.
2. In your settings, on the left hand side click on MCP. You might have to scroll down within the page to see it.
3. Now, scroll to the underside of the page. You must see a bit called Spaces Tools. Within the search bar, type Flux.1-Kontext-Dev and choose the space called black-forest-labs/Flux.1-Kontext-Dev. The page should seem like this after you click on it:

4. For this demo, we’ll use Cursor, but any MCP client should follow an analogous procedure. Scroll back as much as the highest of MCP settings page, and click on on the Cursor icon of the Setup together with your AI assistant section. Now, copy that code snippet and place it in your cursor settings file.

5. Now, once you start a brand new chat session in cursor you may ask it to edit a picture! Note that for now the image should be available via a public URL. You may create a Hugging Face Dataset to store your images online.

Using a preferred public space as a tool may mean you could have to attend longer to receive results. If you happen to visit the space, you may click “Duplicate This Space” to create a non-public version of the space for yourself. If the space is using “ZeroGPU”, you might must update to a PRO account to duplicate it.

6. Bonus: You may as well seek for MCP-compatible spaces with the Hugging Face MCP server! After completing step 4, you too can ask your LLM to search out spaces that accomplish a certain task:

Conclusion

This blog post has walked you thru the exciting latest capabilities that the Model Context Protocol (MCP) brings to Large Language Models. We have seen how Gradio apps, particularly those hosted on Hugging Face Spaces, at the moment are fully MCP compliant, effectively turning Spaces right into a vibrant “App Store” for LLM tools. By connecting these specialized MCP servers, your LLM can transcend basic question-answering and gain powerful latest abilities, from image editing to transcription, to anything you may imagine!