ChatGPT’s New “GPT Mention” Feature Is Here

A brand new and incredibly exciting feature called “GPT Mentions” has just been released in ChatGPT today. GPT Mentions allows users to chat with multiple GPT applications in a single chat window, combining their outputs and chaining their responses together directly in ChatGPT.

This is an extremely interesting addition because it opens up the possibility to leverage different GPT applications’ strengths in creative new ways. Unfortunately, only a limited number of users currently have access to try out the feature. It’s unclear when GPT Mentions will roll out more widely to all ChatGPT users.

In this article, I’ll explain what GPT Mention is all about.

How to use GPT Mention?

GPT mention allows you to inline tag any other custom GPT by using the “@” tag followed by the name of the GPT application. This feature allows switching between different GPTs within the same conversation, taking an agent-like approach to chat.

Check out this new feature in action:

For example, you could start discussing a topic with ChatGPT, then bring in a separate GPT optimized for writing poetry to add a poetic flourish, and then inline tag another GPT app designed for generating images to bring the poem to life. The possibilities are endlessly exciting!

When is it useful?

Here’s an brilliant example of the power of this feature shared by early user Karthik Ramasamy on X:

Using the two GPT apps “Ai PDF” and “Grimoire”, Karthik was able to read and discuss a PDF file using “Ai PDF”, then write sample Python code for it using “Grimoire.”

First, he uploaded the PDF and asked Ai PDF if it contained any Python code for a specific task. Unfortunately there were no Python examples.

He then seamlessly switched to “@Grimoire” within the same ChatGPT conversation, asking it to generate custom sample Python code implementing the desired functionality.

The GPT app wrote the code for him. This is an amazing demonstration of how GPT Mentions enables combining GPTs in ways never before possible to solve problems creatively.

GPT mentions leak

There is currently no official announcement from OpenAI regarding the GPT Mention. It was initially leaked in a GitHub repository chatgpt-source-watch by user Glenn Grant (0xdevalias).

ChatGPT Source Watch is a meticulously curated repository that serves as a treasure trove for those interested in observing the evolution of ChatGPT’s webpack chunks. It gracefully preserves the historical webpack chunks in their original splendor, while also offering a breath of fresh air with unpacked and beautifully formatted versions of the chunk files. This thoughtful touch empowers you to effortlessly analyze the nuances between different builds.

To put it simply, ChatGPT Source Watch is a concise summary of the changes and updates in ChatGPT’s codebase over time.

Some users have found tricks to forcibly enable GPT Mentions by modifying HTTP responses, but this is complex for many. I’d be happy to write a tutorial explaining step-by-step how to try this if there is interest! Please let me know in the comments.

Final Thoughts

In conclusion, I like the idea of having the ability to bring up multiple GPT apps in a single chat instance. However, since the feature is still new and not even rolled out to all users, there are bugs and limitations that need to be worked out.

I hope this feature improves in the future to let us ask one question to multiple GPTs at the same time, instead of mentioning each one separately. That would make things much easier! I’m sure a change like that is coming soon.

‍