“Router” ChatgPT, which automatically chooses the appropriate OPENAI model for work, seems inevitable

Within 2.5 years of the Openai ChatgPT debut, the number of vast language models (LLM), which the company has made available as the power supply options for chatbot hit, is constantly growing.

In fact, there are currently 7 (!!!) Various AI models, which paying ChatgPT subscribers (from USD 20 and more pricey levels) can choose between interaction with trusted chatbot – each with their own strengths and weaknesses.

These include:

1. GPT-4O
2. O3
3. O4-mini
4. O4-Mini-High
5. GPT-4.5 (research preview)
6. GPT-4.1
7. GPT-4.1-Mini

But how should the user decide What Apply their specific prompt, questions or tasks? After all, you can only choose one in time.

Is assist on the way?

It seems that assist is not on the way with OpenAI – because in the last few days there have been reports on X from influential AI, including OPENAI’s own researcher “There (@tzzzl on x) “(Speculated as a member of the Tarun Goginenia technical team) – a modern” router “function, which will automatically choose the best OPENAI model to respond to the user’s input in the flight, depending on the content of the specific input.

As Roon published yesterday on the social network X, July 20, 2025, in response to the statement of the influence of Lisan Al Gaiba that “they do not want a model router that I want to be able to choose used models”:

“You will still be able to choose. It is a product to make sure that the doctors have not got stuck at 4-mini”

Similarly, Yuchen Jin, co -founder and CTO provider’s provider for the summons of AI Hyperbolic laboratories, He wrote in the post x 19 July.

“I heard that the GPT-5 is inevitable, from a diminutive bird.

• This is not one model, but many models. It has a router that switches between reasoning, not justified models and the apply of tools.
• That is why he said that he would “repair the name of the model”: the hints simply automatically lead to the appropriate model.
• GPT-6 trains.

I just hope they don’t delay it in order to obtain further security tests. 🙂“

While probably a much more advanced GPT-5 model would be a huge news if the router can make life much easier and more bright for the average CHATGPT subscriber.

He would also progress on the heels of other third -party products, such as the online token Monster Chatbot, which automatically selects and combines answers from many LLM from third pages to answer user queries.

When asked about the idea for the router and comments “Roon”, the OpenAI spokesman refused to answer or further information.

Solving the problem with excess choice

To make it clear, every time Opeli published modern LLM to the audience, he carefully made available a blog post or release notes or both what this particular model in his opinion is good and designed to assist.

Dedicated to the influence of AI and advanced users can understand very well, in which all these different models are good and not so good.

But ordinary users who do not follow the industry so close or have access to time and finances to test them all on the same input hints and compare the results that they will fight for understanding the stunning range of options.

This may mean that they lack smarter, more bright or more talented answers than chatgpt for their task. And in the case of fields such as medicine, as Roon recalled, the difference can be life or death.

It is also captivating to speculate how the automatic LLM router can change the public in the direction of wider and adopting artificial intelligence.

CHATGPT has already counted 500 million energetic users since March. If more of these people were automatically directed in the direction of more bright and talented LLM to handle AI’s queries, the influence of AI on their loads and the entire global economy seems to be much more edged, creating a positive “snowball” effect.

That is, because more and more people saw more benefits with chatgpt automatically Choosing the right model of artificial intelligence for their queries, and because more enterprises aroused greater performance from this process, more and more people and organizations would probably be convinced by the utility of artificial intelligence and are more willing to pay for it, and as they did, even as they did. more AI drives would spread in the world.

But at the moment it is probably a bit refrained from the fact that the ChatgPT collector requires from the user A. He knows that he even has the choice of models, and B. have a certain level of conscious awareness about what these models are good for. All this is still a hand -powered process.

Like going to a supermarket in your city and staring at the crossing of cereals and various sauces, the average chatgpt user is currently in the face of excess choice.

We hope that every hypothetical Openai router will assist them direct them to the appropriate model product in terms of their needs, when they need it – like a trusted shopkeeper who appeared to free you from the paralysis of the product.

Daily observations in matters of business apply with VB daily

If you want to impress your boss, VB Daily is covered by you. We give you an internal measure about what companies do with generative artificial intelligence, from regulatory changes to practical implementation, so you can share insights for the maximum roi.

Read our Privacy Policy

Thanks for the subscription. Check out more VB newsletter here.

There was a mistake.