How to Make Perplexity Stick to Your Chosen AI Model

Key takeaways:

• New update lets you keep your preferred AI model for the whole chat
• All AI models are now in one easy menu
• Perplexity is listening to user feedback and fixing annoying stuff

So here’s the deal: Perplexity was switching your AI model back to auto mode in the middle of a conversation, and it was driving people nuts. You’d start with GPT-4o, but then your next message would suddenly use auto mode unless you manually switched it back. If you like long chats or need consistent results, that’s a headache.

Now, Perplexity finally heard us. They just rolled out a new UI that makes it way easier to pick your AI model and stick with it. No more random switches, no more digging through menus. I’ll break down what’s changed, why it matters, and how you can use it right now.

The Perplexity Model Switch Problem Was Real

For the past few weeks, a lot of users were complaining about Perplexity’s auto mode suddenly being the default. Even if you started a thread with GPT-4o or another model, any follow-up would jump back to auto mode. You had to keep switching it back, which got old fast.

People on Reddit and Discord were worried Perplexity was having bigger issues—like running out of money or cutting corners for investors. The CEO jumped in and said nah, the company’s fine, they’re just working on making things better and fixing bugs.

The New UI: All AI Models in One Place

The update is simple but makes a big difference. Now, when you go to start a new chat, you’ll see all the available AI models in one menu. No more bouncing between “mode” and “model” menus. Just pick if you want deep research or a regular search, then choose your AI model from the list.

You get options like:

• Best model (usually auto mode)
• Sonar (Perplexity’s own model)
• Claude, GPT, Gemini, Grok
• Even a reasoning section for stuff like D6 R1, OpenAI3 Mini, and Claude 3.7 Sonnet

Once you pick your model, it sticks for the whole chat. There’s even a tiny icon next to your messages showing which model is active. No more surprise switches.

If you want to use deep research, you can still turn that on, and it’ll use the more advanced models for digging deeper.

Why This Update Makes Life Easier

• You don’t have to keep switching models every time you reply
• It’s way more consistent for long conversations
• The menu is cleaner and less confusing
• You always know which AI model is working behind the scenes

If you’re someone who likes to keep your workflow smooth, this is a big relief. For example, if you’re working on a project and need to keep referencing the same info, you want the same AI brain handling your questions.

How to Use the New Model Selector

Here’s how you do it:

1. Head to the Perplexity web app.
2. Start a new thread or chat.
3. Look for the model selector menu—now all in one spot.
4. Pick your AI model.
5. Start chatting. Your model stays locked in for the whole thread.

If you want to switch to deep research for a tougher question, just hit that option. You can always go back to regular mode after.

What If You’re Still Seeing Issues?

This update just dropped, so if you’re still running into weird bugs or if the model switches back unexpectedly, give it a refresh or clear your browser cache. Perplexity says they’re still working on fixing more stuff, so expect more updates soon.

If you’re having trouble with your browser or want to speed things up, check out this guide to making Chrome faster. Sometimes a sluggish browser can mess with how web apps behave.

Perplexity Listening to Feedback—And Why That’s Good

The fact that Perplexity rolled this out so quickly shows they’re actually paying attention to what users want. The CEO is active on Reddit, and the Discord team is dropping updates as soon as new features go live. That’s a good sign for anyone who relies on Perplexity for research, work, or just messing around with different AI models.

If you want to keep your privacy tight while using any online tool, you might also want to manage your website cookies.

Extra Tips for Using Perplexity

• If you’re switching between devices, double-check your model selection when you start a new chat.
• Use the icon next to your messages to keep track of which model is active.
• For more advanced research, try the deep research option—it’s slower but pulls from more sources.

If you’re curious about how to get even more out of Perplexity, or want to try other AI tools, check out how to harness the magic of ChatGPT conversations.

FAQs

How do I make Perplexity stick to my chosen AI model?

Just pick your model from the new selector menu at the start of your chat. It’ll stay on that model for the whole thread unless you change it.

What if my model still switches back to auto mode?

Try refreshing the page or clearing your browser cache. If it keeps happening, it’s probably a bug and should be fixed soon.

Can I access all AI models in one menu now?

Yep, the new update puts all available models in a single dropdown menu for easy access.

What’s the difference between regular and deep research?

Regular search is faster and uses fewer resources. Deep research digs deeper, pulls from more sources, and is better for complex questions.

How do I know which AI model is active in my chat?

Look for the small icon next to your messages—it shows which model is working on your request.