Hugging Face's logo

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
Back to feed
onekq 
posted an update 3 days ago
Post
707
A bigger and harder pain point for reasoning model is to switch modes.

We now have powerful models capable of either system I thinking or system II thinking, but not both, much less switching between the two. But humans can do this quite easily.

ChatGPT and others push the burden to users to switch between models. I guess this is the best we have now.
JLouisBiz

I have a Dmenu script to switch between the models. https://gitea.com/gnusupport/LLM-Helpers/src/branch/main/bin/rcd-llm-dmenu-launher.sh

I just click and then choose the model from the menu.

2025-03-12_21-28.png

You mentioned switching modes and switching model. Now, what do you mean with switching the modes?

And finally, you can just talk to your model and ask it to give you the shell script or any other kind of programming code to help you switch the mode.

The important is that you have defined how to run the model by some command, and then you can put all those commands together in a list, and then you find a way how to switch modes.

I like speaking, even now I'm speaking and getting this comment in text. So that means I could basically speak and have my computer intercept the speech before it comes to any model. And then I can use embeddings to basically recognize if I have given some command. You can even use the simple script recognition. Like you could use string recognition. And then based on your spoken command or maybe the text which you are entering, then you could switch the mode or switch the model.

·
onekq
•
edited about 16 hours ago

Cool! I will check it out.

What I meant by switching is this. Sometimes I'm not satisfied with ChatGPT answer, and realized it needs to think harder. So I switched to o1 and asked again, and most of the times the answer gets better. Then I asked a simple follow-up question which o1 overanalyzed. Then I had to switch back to gpt-4o. I don't actually have the foresight which model fits my question the best. I only know it after I read the answer which is too late.

Now imagine a conversation with a human expert. A human can do such switching remarkably well, hence a cool conversation. This can be actually a metric to read the mileage of an applicant.

In this post