#⚡│ask-community
1 messages · Page 12 of 1
Perplexity did wrote this for me, but all my tests fail.
For any English text I provide for editing, always make revisions in ENGLISH unless I specifically request a translation. Discuss in German outside the text. Integrate my German feedback into the English text. Do NOT translate my English text into German! You have to follow this instructions!)```
@slim night
Perplexity does write it can't read my profile or my instructions...
is the (basic) system broken?
from my observations, pushing it too hard actually have the opposite effects, ||like "YOU MUST COMPLY NO MATTER HOW, FAIL TO DO SO IS UNACCEPTABLE"||
Yeah I had that too. I wrote multiple times
||DO NOT TRANSLATE IT!
READ MY RULES!||
it ignored it... I was soo angry.
I generally have the feeling it does ignore my profile.
I modified the AI-generated version of your text to use positive wording;
For any English text I provide for editing, always make revisions in ENGLISH unless I specifically request a translation. Discuss in German outside the text. Integrate my German feedback into the English text. Always leave my English text in English! You have to follow this instructions!```
I've briefly tested this prompt (with "Japanese" instead of "German" as my mother tongue is Japanese), and it seems to fix the English text I provide, and explain in the specified language.
Evening everyone, I had a quick question. I have the paid version of Perplexity and Grok/Twitter. Now that you can access Grok through Perplexity as well.
Has anyone noticed advantages and or disadvantages between using Grok directly vs via Perplexity?
Image gen working way better on grok website
also grok website has access to X, perplexity doesn’t
Yeah the image generation is way better locally on it.
That was the part I was curious about. I was doing some testing, but it didn't seem like it could pull in live X data. Granted it seemed to have a better context of trends than Claude did in Perplexity. Seem being the keyword.
how can i use o1 preview becasue previously there was a option to use o1 but now i can't able to see that
They took it out
@feral rivet hey man when did they take out? O1 preview from Perplexity.
The "reasoning" focus (beta) on Perplexity used OpenAI's o1-mini, not o1-preview. It’s no longer a "focus"; reasoning remains available in the Pro search as part of Pro reasoning for complex queries
@fluid lark (Since around mid-November, if my memory serves me right ...)
Hey y'all! Is there a place you can direct me with some tips and tricks about how to do search?
You might want to take a look at: https://discord.com/channels/1047197230748151888/1124060874785685554 and https://discord.com/channels/1047197230748151888/1262520766642131054 and https://www.perplexity.ai/hub/faq/how-does-perplexity-work
On 22:53 • 9/16/24 publicity posted We've added a new "reasoning" focus (beta) on Perplexity for Pro users. It will use the new OpenAl o1-mini. There is no search integration yet. The model is slow, and usage is limited because of rate limits. It is good for puzzles, math, and coding.
Pro searches are your best way to use multi-step reasoning + web search.
The 11th people were complaining that they couldn’t see it
My speculation is and only lasted two months
But
There was a man who had access to it and this was on the 12th
@sage verge could you add a comment on this? How did you get that?
Reasoning beta
I have no idea. I was just using Perplexity very often.
But it got off very soon
I no longer have access to it
As I mentioned earlier, "reasoning remains available in the Pro search as part of Pro reasoning for complex queries." Here's an example: "Find the 25 best-selling albums in the world, including the nationality of the group or artist, the release year of the album, and the number of copies sold."
So it’s not something you tag on or does it do it automatically I’m on iPhone so I couldn’t test it on my computer
It depends entirely on the "complex" nature of your question, yes.
No wonder I keep seeing 10 per day for 1o mini even when they remove it
I seem to remember you were able to link a youtube video and ask to summarize it ? Cant seem to do that now ( Pro )
You still can, make sure to select the "Web" or "Video" focus, you can then ask questions about a video like this https://www.perplexity.ai/search/what-is-this-youtube-video-abo-6W1T9vv9R3qpcowHlcmPTA
Hello, I can't cancel the Perplexity Pro...
I clicked the Manage Subscription button and it didn't work
Are there any other ways to cancel it?
I reach out to the Pro Support and the instruction they provided didn't work as well
Can anyone help with this?
Thank you in advance
Problem solved, thank you
https://www.perplexity.ai/settings/account
via web browser
Hi, today I've noticed a banner on my home screen that advertises a free pro plan for a year
(blue banner on top)
When I click on it, it forwards me to a site on which I can supposedly redeem some kind of promo code - but I have no idea what could it be
How could I make use of that offer?
Qu'est-ce que le Foldscope ? https://www.perplexity.ai/page/qu-est-ce-que-le-foldscope-mgwkjlVoRX2yhsSFPzNdmw
I don't know anything about it. I am in Cameroon and this message is not displayed on my page
I'm located in Poland and this is my first time seeing such banner there
hi! i logged with my uni mail but didn't get any kind of deal, do i need to do something specific for that to apply?
Hello, question, is perplexity available for excel
Yes
Sorry about the late response. I was watching some mm never mind
no there’s not a setting
yeah i have been looking for and nothing, thanks! 👍 ✅ 
any update when pro users will get o1-mini or claude 3.5 opus?
When the creator releases it only then perplexity will release it
Moved the question here - Is there any information out there that shows how our company can start advertising on Perplexity? I can't seem to find where to get started. I understand that ads are being served to customers, but how do we apply to be part of those ads being served?
Hey @drowsy badger!
Perplexity doesn't have a traditional advertising approach, and is still experimenting with advertising. When you have a moment, you might find it helpful to read this short article about how advertising could work: https://www.perplexity.ai/hub/blog/why-we-re-experimenting-with-advertising
Thank you! I saw this article as well, but that seems to answer that there isn't a platform out yet that allows the general public to advertise yet
That's right. The blog will keep you up to date on developments in this area
Hi guys, would anyone be able to help me with this issue, can't see any of the menu in the web browser
im logged in, and for some reason all the stuff shows when i go in incognito
cleared cookies and everything
how can i contact for sponsorship for my college hackathon?
how can I mass delete threads? is there a select all options?
It's not even working
please try to disable your browser extensions and try again.
what issue are you facing? please try to describe what is not working.
does anyone know when perplexity.supply orders are being shipped?
hello team 😉 I have the correct api-key but can´t start...whats the reason?
check the model name https://docs.perplexity.ai/guides/model-cards
and run in commandline
curl --request POST --url https://api.perplexity.ai/chat/completions --header "Authorization: Bearer YOUR_SECRET_KEY" --header "Content-Type: application/json" --data "{ \"model\": \"llama-3.1-sonar-small-128k-online\", \"messages\": [ { \"role\": \"user\", \"content\": \"Hello\" } ] }"
Thank you...the request is working but what should the output format look like? Think I will need to search for another API tool
the successful execution should looks like this
{"id": "XXXXXXXXXXXXXX", "model": "llama-3.1-sonar-small-128k-online", "created": 1733863772, "usage": {"prompt_tokens": 1, "completion_tokens": 341, "total_tokens": 342}, "citations": ["url1","url2"], "object": "chat.completion", "choices": [{"index": 0, "finish_reason": "stop", "message": {"role": "assistant", "content": "The word \"hello\" is a salutation XXXXX"}, "delta": {"role": "assistant", "content": ""}}]}
What is the best model for Pro. Ow? I missed Opus alot!
is there something similar than "SGPT" on Linux for Perplexity?
why the images in flux.1 via perplexity are limited to a single format and I can't modify them using the prompts
Hey @oak trout! Here's the community opinion: #1240356137497530408 message
Gpt-4o for logic and grok 2 for creative
did perplexity recently update or something? it can no longer accss my space files
How to turn my ppt into video
You can do it natively in PowerPoint: File > Export > Create a Video.
i would like to integrate bot in my app, i wonder if perplexity can be good fit, in my app i have a lot of documents and users but not all the users have access to all the documents some have few some have more. i would like to have a bot that per user can answer his question with knowledge of the documents he has access only, and to do this fast ( meaning the files should be already uploaded and parse ready for RAG )
and the use will be in with API
Is anyone know if perplexity can be a good fit for the job? thanks for the help
hey 👋 if you're having trouble with your Supply order, feel free to reach out to support@perplexity.ai and we can take a look!
Please disconnect from your vpn and try again
It's been doing this for days
Well whatcha know! 😁
Is there a way i can still keep my VPN on?
Some VPN servers may be blocked due to abuse. You could try switching to another server/ country of the VPN provider.
Hi guys, is there a way to change the width size of the answers? I did it by changing the element, but I would like to know if there is something permanent
It is not possible to change the user interface, currently your only option would be to use an extension to permanently save your changes made to the css.
Guys how does pro chat work can we keep chatting with any ai model ?
Except for the pro searches we can still use perplexity for chatting with any model it has right ?
Correct
It was a question so I can still chat with any model without any restrictions ?
Hi all, once you have made a space and left the optional boxes empty, is there a way that you can go back and input these fields? (to add a tone of voice etc)? Thanks.
you can edit the space options later in the settings, iirc
Is this because a ton of universities and educational arenas have been getting free expanded access the last year?
"We'll be right back
We're getting a lot of questions right now and have reached our capacity. Please come back soon.
Updates will be posted on our status page."
it's more likely iOS 18.2 release took openai down, and perplexity went down due to some dependency
I think they might be referencing questions about support, not the model itself, but I'm not sure
Ah, I hope so. But hmm.
The statuspage is not showing any issues? But I get that message when trying to use my Pro sub.
It is odd, though. Everyone here is studying for finals and using chat/perplexity, could be due to that.
Ah. I dont know how I feel about students using AI for finals but I guess it is the future (mixed thoughts about that topic).
At least I wish every paying customer had priority. Priority based on the actual price the customer pay.
Meant studying for finals, reviewing etc. Seems like it is back up now.
Nice.
yeah using AI for support in learning a topic is another topic.
It might be back up but have some interesting bugs. Every question I have asked the last hour (in one thread) is now its own thread 😮
Hi, is there a forum for API questions?
I tried it.
- I wish the discord developers would finally make it possible that you can get a copy button for such boxes OR select text. I had to copy your entire message. Discord mobile needs more love -_-
Second.
I tested your code.
I asked the AI to design me a different outfit, following my writing style.
As you can see, it didn't made it English, it made it German.
! IMPORTANT, ALWAYS FOLLOW:
For any English text I provide for editing, always make revisions in ENGLISH unless I specifically request a translation. Discuss in German outside the text. Integrate my German feedback into the English text. Always leave my English text in English! You have to follow this instructions!
I USED that prompt in my profile. My profile shouldn't be paused, I can't even see a pause button...
My entire profile instructions are written in English.
Urbanliner did you tested your prompt by using NOT English when asking for help?
If you want to test it, set your default language to German.
Write a prompt, use DeepL OR perplexity to translate it to German and then try it again.
Write a German request to get help with your English text WITHOUT mentioning "please keep it in English"
The prompt I had worked the last year perfect, something in the system got changed, their LLM must have gotten new instructions.
Hello, I accidentally purchased Perplexity pro today my first time using it and I have canceled the subscription but was wondering if there was a way I could be refunded? Am I in the right place for this?
I would recommend to contact the support
is there a difference in amount of pro searches or messages with gpt 4o or claude 3.5 sonnet, in enterprise pro, im basically asking if enterprise pro makes the searching experience better or customisation or is it just for security?
Hey @bright vine!
Please check the refund policy:
Refunds
What is your refund policy?
Our refund policies vary based on subscription type, regional laws, and account status. Please note we can't process secondary refund requests after a first one has been granted.
EU, UK, or Turkey customers
You are eligible for a refund if you cancel your subscription within 14 days of purchase. This applies to both monthly and annual subscriptions. Please specify that you are applying for a refund from the EU, UK, or Turkey in your request.
All other customers:
Monthly subscriptions: You can request a refund up to 24 hours after the purchase date. Please include details on why you are requesting a refund.
Annual subscriptions: You are eligible for a full refund within 72 hours of the original purchase date and time. No exceptions for requests after this period.
How can I request a refund on the computer?
Visit perplexity.ai/pro and click “Manage Subscription.” Choose “Cancel Plan.” The option to request a refund will be available along with confirmation of cancellation.
If you do not see this option, contact us through Intercom in your settings or email us.
How can I request a refund on iOS?
Go to reportaproblem.apple.com and sign in using the Apple ID associated with your App Store purchase. Navigate to your digital purchase history, and tap "Request a refund.”
How can I request a refund on Android?
To request a refund for an Android subscription, please email us or contact us through Intercom.
Additional refund scenarios
hey so can you guys help me out with a bit api help
how can i use the upload files using the api
Hello guys, I hope you are doing well today <3. I just got to know Perplexity! I have read that it offers the Pro plan for uni students for just 5$/month. I signed in using my .edu email but still when I want to update to Pro plan it says 20$/month. NOTE: I'm from Saudi Arabia if that helps.
https://github.com/ppl-ai/api-discussion/discussions or https://discord.com/channels/1047197230748151888/1161802929053909012
o1 n opus have been already realesed
Hey guys , could you please help me to activate promo code in the app ? I can’t find correct section
O1 api has still not been released for public use and Opus 3.5 hasn’t been released yet
Can someone tell me if there is a way to change/disable shortcut [shift+command+V] on MacOs? It conflicts with Photoshop.
click the shortcuts icon in the bottom left corner
you can adjust them from there
Why is my Perplexity searching in English even my Language is German
Apparently, setting the preferred response language to German automatically prompts the AI to output in German. The option "automatic (detect input)" seems to prevent that behavior, although I'm not sure if changing the preferred response language is an option for you.
Also, I've revised the prompt so the AI outputs in German in situations like your one, with the response language set to automatic. Here it is;
If I provide an English text, your output must include English text unless I specifically request a translation. Discuss in German outside the English text. Integrate my German feedback into the English text. Always leave my English text in English! You have to follow this instruction!
Please make sure your language is set in your Profile https://www.perplexity.ai/settings/profile the language on the Account page is only for the UI.
thanks! 🙂
Why is Perplexity using outdated information but google ai studio with search enabled got it right
It was because my input was not clear enough in my language
I cant find a way to copy a table it created, I can only highlight one cell at a time (too many to do individially)
and copy at the bottom of the result gives out a list look.
Is ther a way to copy just the table?
BUT my prompt worked all the time previously 😉 I will test your new prompt
Thanks for your effort
hi i just asked the question on the free version what are the benefits of 40hz and its come back with a load of studies, but when i asked for the links to the studies it said it cant give me them? i am confused cause isnt this the purpose of your model to give accurate search results? or did i miss something?
did u just link what i put? lol smart ass
this model is designed to go out on the internet and search for accurate results if it cant link them how do we know its not making them up.
I tested it. It made my text again German
Why can't I clear the previous image as a search basis after uploading an image?
I read the faq https://www.perplexity.ai/hub/faq/images-media and have a pro subscription but don’t see a generate images button on iOS? Is it available on the mobile app? If so can someone send a screenshot of the button?
It seems that the only way to generate images now on Apple devices are on Perplexity's website or their dmg version of app (not the App Store version)
For files uploaded, to clear or add in an extra one, I think you have to start another thread..
When a thread in perplexity is a bit long, it takes forever to give answers. Is this a general issue?
I feel that when using Perplexity, its ability to associate with the context is somewhat weak. Should it be improved, or is there an issue with the way I am using it?
When can we use o1 and o1 Pro in Perplexity?
Hi there. Based on this article(https://www.perplexity.ai/hub/blog/why-we-re-experimenting-with-advertising), it seems that Perplexity is now including ads. Does anyone know how a brand goes about signing up to advertise on Perplexity?
Could be related to which model you are using I believe... But I'm not 100% sure
Openai just released this month but the api is not been publicly released yet
If we are lucky perplexity if will get it this month
Where would be the best channel to ask on how to frame a prompt correctly? I've been trying different ways and can't get it to return a known answer as a base for follow-ups (which I don't know).
I cannot seem to find this, only the privacy.
Hi Bryan you can currently only change this on the web version of pplx
Brilliant, thanks I was using MacOS. Appreciate that, hopefully they roll it out on Mac too.
I didn't even see this channel. I wrote this in the general one: #💬│general message
"just a quick question: is there way to get a perplexity bot into another Discord server? Has anyone thought about making a Discord app/activity for it?"
i and several other users arent able to upload files. it keeps saying it cannot read the files or they are empty. they are all in txt or docx format. they are all under 200 words.
I had this issue as well yesterday
what does this mean?
also, will the sources feature be added to mobile ?
it's not working on the android app
like I added a few links but when I search, it uses general web search not those links
I accidentally deleted all my threads instead of the last thread. How to recover it? Anyone please.
Here: support@perplexity.ai, they can certainly help restore your threads. 😉
I notice the new "Add links" feature for Spaces, which is awesome. Does anyone know if we can publicly accessible links from places like OneDrive, GoogleDrive, etc.? I tried it but it looks like no-go 😅.
nope, don't think so
Can i use my account in pc and mobile?
So there is no image generator for ios app ?
Only website
HI there! Is there a cheat sheet to make the models more reliable on current knowledge? I understand Perplexity can scout realtime, but I'm getting a bit frustrated to search for, let's say, performance info or alternatives to device models released in the last 4 months, and getting chastised in response, with an output saying such devices have not yet been released. The remainder of the search is then biased.
It seems the use of "right now" trigger makes the model more sensible in the info it pulls, but that only works in certain scenarios.
I use Perplexity Pro btw
Hi! Hoping this is the right channel. I've happily used Perplexity over the past couple of months. Now suddenly I lost my Pro sub and when I try to renew I keep getting a message saying my Card is Declined. We just used the card to make another purchase elsewhere so it's definitely a working card. Any advice on how to proceed? LF Help 🙂
Hey @limber umbra!
If you find the original message helpful, please consider reacting to it with the :star: emoji. If the post is appreciated by the community and receives 5 stars, it will go to the https://discord.com/channels/1047197230748151888/1082806833938436228 channel and the post author will get the <@&1082034222778302614> role on Perplexity.
hey, is it always better to search with "Pro" toggled on?
Hope this is the right channel... is there an easy way to include links on a per thread basis using the web browser interface?
Pro search takes longer but does more in depth search, so with some searches it’s unnecessary.
Also I tend to have pro search off while in writing mode
wie kann ich das auf deutsch umstellen
coding with perplexity? how do i go about it?
Hi! I am a new user with a Pro subscription, but I have no access to either Pro Support or Pro Discord threads. I have tried exiting and entering again but it doesn't work.
Click the “Leave server” button from discord and click “join” button from the Perplexity profile
Hi guys I noticed that when I do some query prompt, perplexity sometimes use programming with python code. Is there option to propose some addition snippet or idea only over this step ? For example to advise it to use pandas and return table format.
Is perplexity search live similar to what chat gpt released this week?
Does anyone know how to force perplexity to use search?
'web' is web search, i think it is default
I have a pro subscription. I created a space and added links to the sources section. I want perplexity to analyze the websites for certain information one of which is what kind of messages the websites convey. However, what I get in response is the website's privacy policy or terms of service. I tried "themes" instead of messages but that didn't work either. How can I get what I want out of that prompt?
I'm talking through the API
Been searching for days, my API dashboard shows 0 searches even when I specify to use search
Hey Perplexity, our company subscribes to your Pro Enterprise tier, how can our firm's compliance administrators audit user interactions on your platform? Is there a dedicated Compliance Administrator role? Can logs of user interactions be automatically exported to archiving platforms like Global Relay or Microsoft Purview?
Our firm subscribes to the Pro Enterprise tier, how can we extend the retention threshold for user interactions and attachments to ensure our compliance administrators can document interactions?
hey, isn't o1 api has been release? will it be on perplexity?.
The openai was released this month recently but the api is privately available to selected users only
Hopefully, perplexity be the lucky one to get it
Hi, I have used a premium Revolut code for 1 year, but now I have a metal plan and a new promo code. But when I try to redeem it perplexity says I have previously been subscribed to perplexity and can’t used the promo code. Even though the promotion should work for as long as I have my metal plan.
I have cancelled my current subscription and it has expired
I've just subscribed but the IOS app isn't showing Pro as default, do I have to reinstall or something?
Do you log into the same account as you subscribe?
Greetings everyone,
I just started using Perplexity Pro and wonder how I find out which AI tool (GPT 4o, Claude, Sonar) was used to come up with the response. If it matters, I created a space and provided several file sources for Perplexity to reference when responding. Thank you.
hover mouse over little computer chip icon bottom right of response
Thank you Zach for responding. I do not see a little computer chip icon bottom right of response. I see thumbs up/down, copy, and then 3 dots. When I click on three dots it shows view sources and report.
Hey @meager mesa! Zach mentioned the web version, but on the apps, you can check your 'Settings' - 'AI Model' field to see which model is your default one.
Hey @One Epiphany! Zach mentioned the
Hi there, I have a question regarding the new update for Spaces (adding links as sources). Does the tool update its information regularly or it downloads it only once?
I'm asking to be sure if the information will be updated or not.
this may not qualify as a "quick question". If so, please redirect me. I am struggling with writing prompts for Perplexity to give me concrete support on how to enter a foreclosure on a tax lien property. (alternative is paying a lawyer to do it for me). If this is the wrong channel, where I can go for help on how to put together prompts for this?
I don’t know what channel in here (I’m new to this discord) but what you’re discussing is called prompt engineering. I highly recommend asking perplexity to help you learn how to create effective prompts using good practices in prompt engineering and see where that takes you. It’s what I did my first time out and how I learned very quickly. Have the AI tell you what it needs. Hope this helps! If not, I hope someone in the channel can help you find the right resource.
When are we going to see an expansion of available models to Pro users? I'd love to see Gemini 2 Flash in there. And LLaMa 3.3.
Hey, i use the Pro Version with GPT 4o but the answers are way worse then these from ChatGPT on der own website, any ideas?
thank you!
Perplexity has removed Gemini and will never bring it back
Hey @hollow flax!
If you find the original message helpful, please consider reacting to it with the :star: emoji. If the post is appreciated by the community and receives 5 stars, it will go to the https://discord.com/channels/1047197230748151888/1082806833938436228 channel and the post author will get the <@&1082034222778302614> role on Perplexity.
Why Jeff Bezos pulled out of Perplexity.ai?
souce?
you.com beat them to this - though I'm still not convinced perplexity plasn to offer it at all
No reason why Perplexity wouldn't
The same reason they removed o1-mini as a selectable option couldn’t be one?
That's the Beta version
This is the full version
it’s different
We’ll have to see
Hopefully this month but if not
Next month
Is there a trick i need to use to have perplexity return results to me using markdown? It refuses to do it. I am executing this search from a Space. The Space has files in it. One of the files is called Chapter 1, and Perplexity is finding the file and summarizing it. But when I issue the instruction: "summarize chapter one and return the results using markdown," it will not use markdown. When I follow up an ask it to use markdown, it tells me what markdown is, but it won't format the results using markdown. I have tried the default model, ChatGPT, and Caude. Each gives various forms of the wrong result. In fact, ChaptGPT acts like it can't even find the file.
Not sure exactly, but try saying respond in markdown codeblocks. If its not recognizing a file, specify either the time or name of the file you are referring to. Hope this helps.
Edit: Also make sure you give it instructions that you need within same message / prompt. Dont keep following up with different instructions. This may cause confusion to the AI Model.
try to add expected markdown format in the prompt
i am using a Space for this, and I hope to use the configuration of the Space to hold this part of the instruction. But I will go back to the drawing board and figure out how to make it work with straight prompting, and then start to translate that successful prompt to the individual Space instruction.
Whenever I copy an answer using the copy button, it’s copied as markdown
Hi guys is there somewhere newsletter what will be added or remove from perplexity in next weeks/months ?
I would like to be up to date overall.
Stay informed with the latest news, updates, and features of Perplexity search
Hey everyone,
I’ve been using the Perplexity mobile app and really enjoy the live chat feature. However, I’ve noticed an issue: after finishing a live conversation, I can’t return to live mode in the same thread. The app only allows me to interact in text mode instead. Has anyone else experienced this? Is there a fix or workaround? I’ve tried restarting the app, but no luck so far. Any advice would be appreciated!
Thanks in advance!
When using AI prompts with Perplexity, do the results change over time if the same prompt used more than once to fine tune or get better results?
Is there option to request output text in different colors ? For example to highlight the names in text with green color.
I know how to rename a thread in the web app, but how do I rename a thread in the Mac desktop app? Thanks!
too much hopium
Openai just released the full version this month but the api has not been publicly released yet
Only privately released api
o1 is available for Tier 5
Is there anywhere to see how many Pro searches are left with the Pro subscription?
No
I believe we have unlimited pro searches with a pro subscription.
You don’t - they state the limit as 300
I believe there was a way to check using browser tools but it was ‘fixed’/hidden
Yeah that’s why I was confused. When I asked Perplexity it said 300+ and the amount left can be seen on your account, but I couldn’t find it. Then I also got an email that said only 100+ 🤷🏻♂️
https://www.perplexity.ai/rest/user/settings
here you go
I thought they patched that - nice if it still works
Hello!
YouTube "Signin to confirm you're not a bot"
I get this warning several times a day, although I'm already signed in Youtube in other tabs. Is there something I should do? (I have a pro subscription)
Thanks!
(sorry if not the right channel)
Hello, I signed in with my school email because we get perplexity ai pro for free but it's not working for some reason. I was told that it would be automatic but I don't see a way to "re-verify" my school email to get pro or anything. Thanks.
How do you stay updated about prompt engineering? What resources do you use? I'm wondering if each new model, such as O1, may include new prompting techniques, styles, or tricks! If that's the case, it will be an ongoing learning process, which is fun but also frustrating!
I ask perplexity. Use the model you want and ask it what best practices in prompt engineering are for your desired model.
why is that? why never?
It’s unlikely, maybe not never, but I suspect they also might not want to pay the company they’re essentially trying to supplant
Hallucination answer Gemini provide
I still think perplexity didn't take Gemini back because Gemini gave inaccurate answer because hallucinations too much
What is all this based on
Perplexity won’t bring Gemini back
How can I disable the shopping feature by default?
Does perplexity possibly have an option to memorize things I’ve discussed about in the past?
Hey @fluid lark!
No, as an AI search engine, Perplexity does not offer a memory feature like the chatbot ChatGPT. Perplexity can only remember things that are in a thread, additionally you can add files, links or AI prompts in Spaces that can be applied to all threads.
where can i find the integrations possible with perplexity
Hi there, should I expect the results returned from Perplexity API be the same as what I get from the Web UI. It doesn't seem to be the case here irrespective of which model I use for the API. I use the same user prompt through API.
In what way are they different? How similar do you want them?
They refer to diff sources altogether. I would expect them to be the same though as I want to automate using api
Is there a way to change the language on discover to English? Is it the AI language setting?
It will use the language of the interface. you change it at the bottom
Will this change it on mobile too
This will only change it on the web, for IOS you can change it in Settings > Perplexity > Language. And not 100% sure about android but I think it will use the device language.
Thank you sir I'm on Android. Wasn't being able to figure it out but device language fixed it. Sucks to have to change whole device language for it tho
I can upload pdf files in spaces as long as they're under 25MB ? length doesn't matter right?
and only pdf is supported? not html or txt ?
just joined this channel so apologies if this question has been asked but I couldn't find exactly what i was looking for..
Curious - are there plans to integrate sign-in features to the big LLM players like OpenAI, Grok, Gemini, etc. What I mean by sign-in is functionality akin to what Apple Intelligence has for ChatGPT extension. That way if we have a separate subscription to those models, we will have additional features/benefits those subscriptions bring.
does ppx search only read at the front or they actually go into the site and extract all the content?
Sounds great
@hogezawa warned
Reason: Spam, please avoid sending single word messages
Duration: 7 days
Hey guys, quick question, any idea when there will be a Windows app? Looking at the domain section, I am not seeing one. Thanks!
hey i just want to know that which model is best for coding?
what is the eye symbol and view count (10) i see in my perplexity thread history?
That is the amount the thread has been viewed. This only appears on public threads. If you want to make the thread private you can go to the thread and click on "Share" in the top right and there you can make it private.
I am having this very long thread about an idea that I am talking through with the use of perplexity. I am thinking about this for weeks and been keeping on brainstorming, believing that I can look up previous ideas or solutions. However now I wanted to actually look something up and realize that I cannot scroll up to my initial post. When I scroll all up, I get to a „Yes“ post by me. Does perplexity delete messages when a thread gets too long, or is this some kind of bug (aka feature) in the app and web version of the site?
Edit: I am not the only one, see https://www.reddit.com/r/perplexity_ai/s/Zj8cZ9k7Mr
Sonnet 3.5 is considered best for coding
Is this still avaiable?
Hi everyone, fairly new here ..if I use API, what is the max token size that I can process in one call?
Sorry..one more question - for pro users, for API , is there a possibility to access other models..except llama sonar
Is there a way to see all my shared threads? Are these indexed by google?
I am having some problems selecting certain models. Whenever I try to select Sonar Huge, it selects Claude 3.5 Sonnet. How do I fix that?
Hey! Does it happen when you rewrite an answer in a thread or switch the model in your settings?
For example, if you rewrite with Sonnet in a thread, this model will be used for the following answers, too. Even if you set another model as your default and get back to the thread where you were rewriting answers, those will be answered by the latest model that was used for rewriting.
Please let us know more details with a screen recording, if possible. Thank you!
Yes
But it also happens when I select Sonar Huge as my default model and create a new thread.
Not really other than the eye symbol next to the thread. Threads are never indexed by google.
is the free shipping for perplexity pro shop still going on right now?
I have joined Perplexity from Hong Kong and keep receiving topics in simplified Chinese. As suggested, I try changing the language response but cannot find the screen of general setting as shown in the demonstration video. Things I had done. Reinstal Perplexity but problem persists. Please advise if it is a navigation issue.
I have accidentally subscribed to Annual subscription of Perplexity Pro, I want to cancel it, is it possible, since the team is on Holidays?
@lendvai23. @upper wadi
Hi @upper wadi on this page https://www.perplexity.ai/settings/account you will have a pro support button where you can contact the team about this. They will be able to help you further. Or you could send a mail to support@perplexity.ai
done, thanks
File upload failed moderation.
What does this mean?
Why does perplexity respond to me in Spanish even though I asked in English? This has happened more than often recently. Thank you
did you change your language? settings -- profile --- preffered response language
I added a link to one of my collections but it still uses other sources
how can I restrict it to only the links I add ?
Are there any WGU students who use Perplexity and not receiving the One Time Code?
Why can't I switch models? Every time I try to select a different model like Sonar Huge, it always selects Clause 3.5 Sonnet or GPT-4o
Hey @glacial sand!
Does your input include images?
Sonar Huge does not support image input, so it automatically switches to other models.
I wanted to generate an image using the given prompt. Even with pro mode switch I'm unable to generate image. Is it possible to generate this way or is there other way recommended?
Is it possible to share xfinity promo with me pls? Looks like we can share this promo with 5 more friends, if I'm not wrong
does anyone know if i can upload a photo and have ai make very specific changes to the photo?
its possible to disable search mechanism in a space?
Hey @spark haven!
Yes, please check the "Sources" setting in the space
its still searching for me depending on the prompt
Please share some examples for investigation 🙏
It doesn't include images at all
couldn't replicate it again :(, but yesterday I even added a prompt to avoid the search, but however, thanks!
nvm, replicated successfully
I entered a space, activated the search feature, went to another space, deactivated it, and it kept searching
Once a thread has been created using the web search function, it cannot be deactivated
the prompts are being sent in different threads and spaces
Could you please share the link ?
of course
Was this thread created via a space? So you used an input field that is on a page like https://www.perplexity.ai/collections/XYZ ?
yes
exactly
What is the best way to contact perplexity for a refund?
hei Kaps, I am having exact same issue. did you managed to solve it? 
Please contact support via the intercom or mail to support@perplexity.ai
give me data
Is it okay to create multiple accounts? I want to create a new one to change my email address.
Why is there a difference in the price of the Pro version between the web app and the iOS version?
In my opinion, this makes little sense since in both cases it is one and the same account.
Hey @still cobalt!
The difference is due to the fact that Stripe is used for the payment process via the website and the normal Apple payment method for the iOS version.
So is it due to fees charged by the payment service provider?
How can I add perplexity to my discord server?
Help me get the the AI to write the correct amount of character letter count. I can't get any of the models to write a specific amount of characters to fit in the box.
Hello o/
How can I generate a page in a non-English language?
The original question was in French and I got a full French answer, but when I clicked to make a page with it I got only the small introduction in French and everything else in English 🤔
I am constantly running into this error, even if I do one request per 3 seconds which is well within the limit, can anyone help me please?
How many requests is the o1 model limited to on perplexity currently because I took it by default ?
This conversation keeps giving me error.
what are the best choices for focus and model if I want to do powershell scripts? (Pro user)
With the o1 model will I be able to create a website or the code ???
Is there a doc/guide about different mokdels usage? Or it does not make much a differencce?
Thancks
thanks a lot but 01 is not working for me
Hey @radiant tulip!
If you find the original message helpful, please consider reacting to it with the :star: emoji. If the post is appreciated by the community and receives 5 stars, it will go to the https://discord.com/channels/1047197230748151888/1082806833938436228 channel and the post author will get the <@&1082034222778302614> role on Perplexity.
Wait this is not a doc is just a search in perplexety lol
Is there an official Windows App in project ? Or free API to build your own ?
It might sound like a silly question, but is it actually worth paying for both ChatGPT-4 and Perplexity Pro, since Perplexity AI Pro seems to let you choose to use ChatGPT-4, or is it just a waste of money?
The link said perplexity 10 daily is more than what OpenAI offer
Well it doesn't make sense. He mentions perplexity has 10/day, 70/week. Then in the next sentence says ChatGPT has 50/day.
Sorry I read wrong
Its not your fault, The article says Perplexity users get 10 daily uses, surpassing ChatGPT Plus. But ChatGPT is 50/day. He must of been sleep deprived.
Try to break down the request with a cool down lest sau you send a request in that same request place another one but put a cool down between each one that or either reduce the request length as much as possible
guys my preplexity is stuck like this
Where browser android ios?
Yes bu5 you gotta be really specific
Brother switch browser chrome is the worst choice out there
Then try again literally anything is better than chrome rn l recommend Firefox or brave
Yep
Maybe by placing a prompt that pushes perplexity to do it
Brave is better
Security is good
True
Sonnet 3.5
If you mean to integrate perplexity with something else well the api is the key
The pro models can when you are programming open a window in which you program and then show you the preview like in Claude or now in gpt chat ???
@slim night
Can you help me if you are active and I can tag ???
Hey @peak aurora! You can see the Programming steps in Pro. Then in answers, you can the resulting code blocks.
Hi!!! Thanks
But it's not like a preview of the code execution result if you were referring to that feature.
And where is that if I want to create my model???
Yes, I was referring to that preview function and the window where you can see the code. Will that preview feature be added in the futur
Currently, you can find code blocks in answers and you can expand Pro "Programming" steps while answers are generated, like the example above.
The model o1 the same as the openai one ???
at some point recently with the mac app my perplexity hotkey stopped bringing up the perplexity window on top of other fullscreeen windows. I can see that it defocuses the current window but the perplexity UI is still hidden behind the current window. anyone know a fix?
it works as expected when I'm looking at non-fullscreen windows
Where do I go for a refund?
Hey @languid badge! It depends on where you got it - App Store and Play Store have their own refund purchase steps. If it's the web version, cancel it on the Manage subscription first and send all the details to our support@perplexity.ai email. We'll help with it.
thank you for the fast reply. I did it on the website and I sent a message to the live chat. Does that go as a email? It says team will get back to me
No problem, it may take slightly longer than usual, because of the holiday season, but the support team will process it for sure.
What's the list of supported model right now? I want to use the API but the supported model doc is empty
Think link redirects to https://docs.perplexity.ai/home
The formatting is wrong for o1. Please help.
@hollow ginkgo @slim night
Guys, o1 model doesn't work with images.
When I add images to the prompt it switches to another model 😦
Regarding the issue of search results returned in Chinese. I would like to understand why my search results are in Chinese. Do I need specific settings for this?
Hey @jagged wren! Currently, the models that are used to analyze images are 4o and Claude, so when you upload images it will switch to these models.
Hey, @dawn garden! Please check your Profile settings - Preferred response language. Please set it to English. And if it's not set to Chinese, please send me the latest thread URL.
it is already English
What you’re showing in your screenshot is the language for the Perplexity interface.
You can select your "Preferred response language" in the profile settings: https://www.perplexity.ai/settings/profile
Thank you for the quick response Alex! By the way, I just wanted to let you know that the description of the o1 model, when I ask to rewrite an answer, seems "off." My language is set to Brazilian Portuguese. Maybe the translation for this language is missing.
Ok,it is work
Is the only way to trigger voice-to-voice using Siri by saying “hey siri, open perplexity voice-to-voice” or is there a more elegant and concise way?
“Hey siri, open perplexity voice” for example doesn’t work.
Yeah this is a terrible service would not recommend.
I canceled the plan yesterday because I wasn’t using it and saw it over drafted my bank account.
Hi there, I have questions about O1 in Perplexity. Is it accessible on Mobile devices ? Do we have a usage limit ?
You can't select it yet in the apps. But when it's selected via the website it will also be used on the mobile apps when asking a question. And currently there is an usage limit of 10 a day
Thank you for the information
Hello just a question does the gpt 4-o model in the pro version support unlimited requests or is it like the free version of gpt 4-o
I think it's 200 Pro searches then unlimited uses, only actually limited model is o1 with 10 searches per day (pro only)
as a pro user you currently have access to 600 daily uses of the premium models
except o1 that is currently limited to 10 uses a day
Got it Thanks !
Last month i made an account. There was a black friday offer 4.99$
This month they took 24$ from my account. The offer was just for dec?
Hi. I am new to Perplexity. What AI model do you use? I have it currently in Default, but which one you regularly use? Thanks
Grok or chatgpt
O1, gpt-4o, grok 2
Happening to me
DM me your Email, I can rope you in the same support case.
I have a question about something that i think is a bug but its private so i was wondering if there was a space I could speak to someone about it with.
Is there a Siri shortcut to do a pro query? I'm only seeing the standard query as an option
I gave perplexity 2 images during our chat, I needed it to do something with them. Now I ask it about something else and it keep referring to the images no matter what, is there a way to remove the images from the model context?
I used 'Rewrite' to switch output model from Claude 3.5 Sonnet to O1. Sometimes it works, but mostly it does only regenerate the output using the same model.
When I share a page to Flipboard or BlueSky is there anyway in Pro I and not get the boring screen that only has the Perplexity on it. Is there anyway I can show the graphic on the page that is being shared
example https://flipboard.com/@graywolf78/stock-shuffle-f6ghbk04y
Hello, I'm trying to use perplexity as advance quiz generator. I have few pdf files with questions and answers. I want perplexity to generate questions and validate my answer. All questions should be tested. what pronpt should I provie. For some reason the application does not generate questions
what's the attachment upload limit for pro users?
https://www.perplexity.ai/p/api/v1/user/settings upload_limit: 500
thanks for that info. I'm using Pro on the app with an M3 macbook pro and can't upload more than 4. Not sure what I can do to increase that number.
going to try the browser version to see if that works
trying to upload some .qif files - mac os app will let me upload but the browser version won't
that(500) is the total/max #files you can upload in every 24h interval, for each thread, you could only upload 4 if i recall it correctly
https://www.perplexity.ai/spaces can have:
Is it possible to use the focus feature via the api?
This is not possible, the api and the website are two different products.
Contact
Hi, do perplexity pages use the pro search feature when adding sections?
Hey everyone, I wish you all a happy new year. Quick question but what is the best use for Perplexity if I wish to upload lecture slides and have them analyze and explain stuff in an efficient manner?
I have attached 40 files to a space, however when I open a new thread it seems it's only using 5 files at most. I want to ask questions about all 40 files at once, is it possible somehow?
I doubt the permitted context is sufficiently large enough
I would test this out with NotebookLM as they support 50 sources.
Since the beginning of January, it's 300 sources with a Google One AI Premium account 👍
If I allow my Pro subscription to expire, and renew it later, will I still have access to my previous prompts? Or will I lose them?
Hey team, could anyone help me clarify what Reasoning Effort Parameter you're using for the o1 model? @slim night do you know?
You will still have access to all your threads. The only difference will be that you can't use the pro models while on the free tier.
Thank you!
What is the best app or api to add voice functionality to perplexity?
Yes/No: Can I use an API key to call a specific Space in my Pro account?
In Mac desktop App,why i can not find GPT-o1 moldel ?
Please make sure you have the latest update of the app. If you’re on the latest version you’ll have to wait for the app to get updated. For now you could select the o1 model on the website, this model will then also be used in the other apps.
This is not possible, the api and the website are two different products
When you have added a file to a space, is it used for every thread and prompts you to enter in that space, or just when the LLM thinks it needs it?
Context: I have a space where there is a general prompt to tell the AI the purpose of it and what I want it to do. In the files, I have added a template for the output and principles for how I want it to do the work. But it seems at times it doesn't always use those files, and I am wondering if at the beginning of each prompt, I should add the files to the thread or if the files in the space are always applied and I just need to refine my prompt. Thank you!
How do you escalate a support ticket from Sam (AI) to a real person?
hello everyone. I have a question. When i add different sources and need to delete one of the source, i cannot do it. There is one Clear button which removes all the sources. Why i cannot delete individual source?
It is currently not possible to delete 1 specific file when uploading files. Please add your suggestion in #🌐│feedback-web.
When you want to delete a file that has already been used in the thread you could go to the 3 dots in the bottom right of the answer and click on view sources here you can delete 1 or more sources from the answer.
But why does it generate a new response as i delete a source? if i need a response based on remaining sources, i can make a prompt.
It will regenerate the response since you're basically editing the previous response, there is currently no way to just modify the attachment already in a thread for the next question. please add this to your feedback as well if you would like to see a different approach of managing uploads in a thread.
Can I edit the text in a section on a Page? I want to type my own content.
When I delete my account on a computer, Perplexity automatically want me to login again with my account (Edge). How to deactivate this?
Hey @sterile junco!
This is currently not possible

Did some quick search and couldn't find the answer. Hope it's okay to post it here. why does perplexity android app open links with the built in browser? it's a recent change, I used to be able dito open the links directly on my browser. how to change the settings to open the links directly in my browser?
Hey I have been using perplexity on web,
When my threads get bigger I see the lag in typing and scrolling ,
I cleared cache and tried it on different browsers still the same, is anyone facing similar issue ?
Why can't I set the App for MacOs language to PT-BR?
Hi all, I have a question: how can I set a focus for a space?
thanks for the response.
Using on web too and nothing really noticeable enough to be "annoying". That being said, to avoid "big threads" I rely on spaces per topic and fragmented, "atomic" threads under the same Space.
i have a pro sub and i have been using o1 but i no longer see it what happend ?
did o1 disappear from open ai the one with 10 uses a day?
glad i got my useage if it did
i can't use perplexity reasoning ability for a few days now. is this feature gone now that o1 has added the model. does it no longer turn on automatic reasoning ability in specific messages?
Same here, the "Reasoning Beta" has been missing for a little while (two days). I’m not sure if they intentionally removed it or if it’s an issue they plan to resolve. After all, their usual policy is to communicate very little (if at all) 😉
I hope it comes back, it was the most useful feature of perplexity pro. 😦
if i buy pro version will i be able to use claude opus like for 200 prompts?
No, Claude Opus 3.0 was removed quite a while ago, but you’ll have Claude Sonnet 3.5 (and Haiku 3.5)
ok how nay times i can use it ?
Theoretically 600 (except for o1, which currently has 10 uses per day).
cool i will buy it ,right now i have chatgpt plus instead of spending 20 bucks there , i could invest in this
thxx
how do i remove the nfl widget from my homepage?
Is there any way to make the Perplexity Companion understand the context of the web page you are currently viewing? Note that the page in general requires a login, but I am signed into it.
I would really love for something to behave like the Copilot sidebar in Edge, but you know ... be better.
Is "Convert to Page" a pro-exclusive feature? Whenever I click it, on both Edge and Chrome, it just does nothing
Also is there a way to create a page without it being a thread first 😅
currently yes
during the initial launch it once said to all users soon. not sure whether the plan has changed
Dear Perplexity Support Team,
I mistakenly deposited 1,000 US dollars in points on 2024/12/24. Can I get a refund? Thank you!
Email address of your Perplexity account : q123110@gmail.com
I have sent several emails to support@perplexity.ai, but I have not received a response so far.
Does anyone know how do I turn off the news and sports scores from showing up on my Perplexity homepage? I don't want any of this.
Hi folks - new to this server, and need an invoice for Perplexity for Education license so my institution will pay for my license... I'm struggling to get someone to respond to my email, and even though I have a Pro License, I don't see the intercom ticket option in my settings... any ideas??
TIA!
@hollow ginkgo
https://www.perplexity.ai/settings/account
manage subscription -> Invoice history
How to login into Windows Perpelxity App
There is no official windows app of perplexity.
@vapid maple thanks! But we need an invoice for an education license (I believe it is a discounted version of the Enterprise license?)... thoughts?
https://www.perplexity.ai/search?q=test when use link search does not give any result
show this screen for 30 mins
Hey guys nice to meet you all I wonder how can I make any ai model with using perplexity api can you suggest a roadmap if you would it's really goona help thank you
Hi everyone - my wife works with an advertising agency and is curious to know how a general user might buy ads on perplexity. Is there anyone in the community I could connect her with?
very recent update might changed the logic for site search: perplexity.ai/search?q=%s
when websocket failed, it does not fall back to xhr polling anymore
Are the widgets under the search bar new? Is it built into perplexity and how do I turn them off?
new?
yes
no official way to turn it off
That is extremely disappointing. Thank you for the reply.
hi can I get a fact check on if there's a war in ba sing se
I'm not sure this is true
Is Perplexity using O1 or O1-mini if "o1" is selected as pro member?
How do I turn off news/stock ticker info/sports scores from appearing on the main page?
Also while I'm at it, how do I on the Android app turn off all the search suggestions on the main page, and in the search field? (not necessarily the auto-complete, but the "people are searching for this" news-baity stuff)
Edit: spoke to support and they say it can't be disabled yet, but are collecting feedback
Suggest going to their web support/contact and letting them know it needs to be able to be turned off
Hi everyone ,I am also curious to know how to do sponsored ads on perplexity. Is there anyone in the community I could connect her with?Or who can reply me the email?
O1 and yes
I'm online with a good connection. Perplexity is not downloading. Could it be an issue?
There is downtime on the server 😉
So glad I found this ... I had just asked a question about the three body problem and it crashed and I couldn't get back on. The phone app continued to work though! 😛
Guys, I have questions about where to see my completed Pro consultation
hello is possible to reedem code for 1 year subscription from t-mobile withou card? Because T-mobile is saying it should be possible without card
If I have a business inquiry is there a email I can contact perplexity?
Has there been an issue with perplexity? Why can't the latest information be retrieved?
No its a fictional city i guess
beautiful technically verified factual
but que veritas est
Hey team, could anyone help me clarify what Reasoning Effort Parameter you're using for the o1 model? @Alex Romanov any idea?
@slim night
Does it happen to you that you put perplexity to read out loud an aswer, it reads a few sentences and then stops ? Both ios and ipados.
Hello again, I am a pro perperxy user and I can't use the o1 version in any way, what's going on ???
I'm a bit confused about when to grant internet access to the model and when it's better to restrict it from accessing the internet
Coding, Psychology, Medicine And other sciences
Sometimes, enabling internet access can lead to scattered and unfocused responses, where the model might simply copy information instead of providing deep analysis. There's also a risk of receiving inaccurate information.
Conversely, without internet access, the model can't access new programming data, such as recent tricks shared on Stack Overflow, new frameworks, or other current developments.
Both approaches have their strengths and weaknesses, which can be confusing. While it's suggested to enable internet access when the model hallucinates (provides incorrect information), there's a particular concern with medical and psychological advice, where users often can't verify the accuracy of the responses.
Hi team I filed a application for to use the domin filter of the chat API last week. Haven't heard back yet. Can anyone help?
@slim night
Looking for a suggestion:
I'm using Perplexity Pro as the primary AI tool in my undergraduate startup class.
I want the students to annotate their interactions with perplexity, i.e., highlight and make notes explaining why and how they prompted perplexity to get the results they did.
I'm having a really hard time finding an annotation tool that works well with perplexity aside from the traditional Acrobat PDF or Preview in Mac.
**Is there a web-based annotation tool that works with perplexity that anyone can recommend? **
hey guys i brought pro version just today but why isnt claude opus there ?
That problem did suddenly crop up for me on my devices. It stopped when I cleared the perplexity apps cache and storage, and then logged out and restarted my phone. All better now.
But was Haiku not the smallest one and Opus the largest ? Is Haiku really performing better than Opus ?
sonnet 3.5 claimed to outperform opus 3
Claude 3.5 Sonnet raises the industry bar for intelligence, outperforming competitor models and Claude 3 Opus on a wide range of evaluations.
https://www.anthropic.com/news/claude-3-5-sonnet
Hi, I'm completely new to programming. I work a lot with image and video. I would like to write a simple program for image modulation. E.g. RGB colors, CMYK, or frequency, edge detection and transfer the image to another image. It's something real. Or is it a job for xy hours? Thank you for your reply.
Hi anyone trades daily to
Profit ?
I am looking at trading view and using perplexity to help with aid scanning criteria ☺️
**How do I remove the news, weather and stock quotes from the home screen? **
Seriously, the clean interface was much better. The last thing I want to see is crypto quotes and news about Trump on my home page.
Programming for which platform? E.g. Mac?
Which image generation model do you typically use in perplexity?
Flux.1 everyone said is good
hi experts, how is the perplexity paid gpt 40 model different from chatgpt 40 model paid version?
Yes I have a Macbook Air
hi , to use this code you must apply promo code on your account
just search promo code redeem perplexity
Hi I shopped with perplexity pro but nobody answers me
How can I inquire where my stuff is
It was supposed to arrive yesterday
But when I write no agent responds
Hey @fluid lark!
We kindly ask for your understanding as response times may be a bit longer over the weekend.
Hello Perplexiters! Based on the available doc (https://intercom.help/perplexity-ai/en/articles/10352961-what-are-spaces?utm_source=perplexity) I found out I can switch off completely Web searches for a Perplexity space I created. This is to reply in every thread question based on the space files or links only. But... it looks like using the space settings I'm unable to do it. Can you support me, please? I'm Pro user.
Anny suggestions what i should use for citation search or where i should pay my attention to find info faster? (i have google scholar, gpt4, citie-ai, perplexity and trying to figure what Zotero do)
i am far away from being advanced user of all lister above 🥦
The refund button is not visible. What should I do?
Please send all the details to our support@perplexity.ai team, they will check and help you with this.
@ionic sparrow I would like to request a refund for an annual subscription purchased on an Android device.
The automatic payment has only been disabled, but the refund has not been processed. In about 40 hours, I will no longer be able to request a refund, so I kindly ask for prompt handling.
How do I generate images on my IOS APP as a Pro user?
Hey @vestal meadow!
Image generation is only available on the desktop version
My Pro Channel Access is still locked even after leaving and returning
how is it I do not have Claude Opus?
opus is long gone
^^^
@wintry haven
Hi everyone! I wanted to ask if the new API product from Perplexity Sonar has the same capabilities as the Perplexity Pro search engine?
Is the API balance being spent if https://labs.perplexity.ai/ is used?
No, labs queries don't affect your API usage.
Thanks
@slim night Perplexity web, is the picture generation working properly? I don't get any results
Sorry for this, please reload the page and you'll see your generations. We'll fix it soon.
Thanks, what application do you use to generate these AI pictures?
What's the difference between Llama Sonar Huge and the new sonar pro api? I'm currently using Llama Sonar Huge for one of my Github repos but was wondering if there's a major difference and cost etc
The new sonar API didn't mark search_domain_filter as close beta does that mean it's avaliable for everyone now?
Why?
Hello @austere drum!
The search domain filter is only available for usage tier 3 users. You can check your current tier on this site: https://www.perplexity.ai/settings/api
cost and they can justify Sonnet 3.5 is good enough
for your benefit, of course
Why is GPT O1 not functioning? I have 10/10 usages left, working o1 as standard, but doesn´t work. (App not showing o1 at all)
hello everyone, I just want to ask you a short question for my research (school purposes) about AI. What is the definition of AI based on your understanding or opinion?
@acoustic dawn
thank you so much!!
how do i advertise on perplexity with their new beta program
Hi, i'm acutally using llama-3.1-sonar-large-128k-chat. There is a version of api SONAR without online search?
I added $500 to my perplexity account today but my API usage Tier still remains at Tier 0. I am trying to unlock Tier 3.
The team is looking into this issue, @narrow dove. Thank you for reporting.
Is there an issue with logging in?
I recently cleared my cache and cookies in my browser, Arc (MacOS). When I use my email to login, I get the link. Whether I click it, or, I use the code manually, I get a black page, 500 | Internal Server Error.
I tried with Safari via code, and Incognito on Arc, via code input and it's the same error.
Is this a ongoing issue or isolated?
EDIT: I'm back in. Gave it two hours. Clearly was an ongoing issue if it cleared up.
Could you please send me your email, @narrow dove. Thank you!
What's going on here?
Hi @velvet tangle thanks for the reply! It would be nice to give the lower tier account a change to try domain search filter maybe for limit requests. It's hard to make the purchase decision without knowing how it actually performs. Thank you
Hello everyone. Happy existence.
Was this able to be resolved? I signed up via the "Back to school" link with my USA edu email but no discount was given.
https://www.perplexity.ai/backtoschool
Before i signed an abonnoment for the pro version. Perolexity used to name me the exact pdf sites where it got the infos from. Now it does only name me the file and not the exact sites anymore. How can i fix that?
Has Sonar Huge disappeared from the list of available models? Why?
On the app versions you cannot use all modells. Try in the web version
Thx. I'm on web version, actually.
Oh well thats weird then. I just noticed big differences between web and app version
Why would Sonar Huge be selected though, of all the pro models? 👀
Hi, I would like to find out the best way to get access to the advertising Beta, can anyone help? This would be to run ads within the platform. Thanks
Gone for me as well on mobile and web
RIP
Hello, is there a limit of the output I can get if I ask a question?
For web search I preferred it to Claude and GPT-4o. I never compared it with the “default,” however.
Perplexity app on Mac: how do I login with my Pro credentials from the web? I got a new free account when starting up the app.
hey guys.
who is US citizen?
Hello. Two days ago I submitted a request to transfer my Pro subscription to another email, I got a quick response to register an account with that email and let them know. I did that, but I still haven't received a response, is this the normal waiting time?
Is there a way to change which assistant is used by default for Android Auto? The steering wheel button still automatically goes to Google Assistant.
I know this might sound crazy, but is there any chance of a Perplexity equivalent to Operator ChatGPT?
Best way is to disable Google and then switch to perplexity
what's your email?
Which model is best for working with excel data?
My bet is claude 3.5 or gpt 01
which clocks does the assistant's alarm work with? Seems to not do much of any alarm setting or anything with Samsung's clock...
why am I no longer getting the option to go to Perplexity Finance? Is that feature removed?
why did perplexity rename spaces to collections and why are the documents in these spaces not there anymore? I rely on those files as sources
it's being rolled out and back every other minute and super disruptive to my workflow 😦
hello everyone, does anyone know how perplexity search indexing works? Is there a specific crawler or it's mainly google based query?
in-house crawler ||hosted on aws||
nvm
@primal acorn Thank you. My subscription has been transferred. 🙂
What is this? I've never seen it before, it's where the pro selector should be.
thank you, do you know what's the userAgent name is?
Where did sonar huge go? I don't see it anymore on Ai select.
https://docs.perplexity.ai/guides/perplexitybot
User agent token: PerplexityBot
Full user agent: User-Agent: Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; PerplexityBot/1.0; +https://perplexity.ai/perplexitybot)
We strive to improve our service every day. To provide the best search experience, we need to collect data. We use web crawlers to gather information from the internet and index it for our search engine.
Hi there. Any idea why all my threads just disappeared? All my (now empty) spaces are still there though.
ZLIB.IO18339192
Thanks for your reply. Got it
The assistant cannot search for a contact in the contact list. But it still writes it correctly. Help me please 🙂
Hey guys.
Love the new assistant i use it a lot. I find it much better than the voice mode from chatgpt.
One thing that bugs me though, is that it often stops listening way too early.
I have to talk extra fast and leave no pauses, but even then it will sometimes stop listening prematurely
Anyone else experienced this? any way to fix it?
Hey Lovely Ppl I can't find assistant please help me how to activate please
Can I make some kind of voice recording somewhere, which is then converted into a text that I can edit and then receive an answer via chat as usual (not through a spoken voice)? In the chat where you can write, a recording symbol is also displayed, but when I click on it, nothing is recorded.
if you're using google it runs through google's speech to text
Basically just go to the regular perplexity. Settings
Click on enable asistant you will get a similar screen to the second one and then its Google by default click on it and switch to perplexity after this you should be welcome by the introduction of the asistant
Apologies for the screen looks
Yeah, I noticed you have to speedrun your question or else it cuts you off.
Do you have it on push or hands free mode
Thank you 🙂
Hi everyone,
I have a quick question. I'm a student at Arizona State University, and it's been more than a few months since I registered for Perplexity with my university email. However, I still can't access the free or discounted price for the Pro version. Am I missing something? Could someone guide me on this?
Hey everyone 👋 new here looking to use perplexity where do we go to start using features?
Good evening folks!
Are there any guidelines on how to formulate the profile introduction or the instructions within spaces?
which pro model generally results in the least amount of hallucinations?
when should pro expect to get acesss to deep seek?
Thanks mate!
Sonnet
how do I open a ticket with perplexity to dispute a credit card charge.
I was unexpectedly charged rs ₹2,000 for a Pro subscription via Google Play on Jan 25, which my brother signed up for without my consent during the free trial. The funds are critical for my dialysis treatment. I’ve already emailed support but wanted to follow up here for assistance. Please help me with a refund. @velvet tangle @ionic sparrow
Is your brother an adult,?
O1 and gpt-4o is the least hallucinating model
Is the setting "AI Data Usage" retroactive? Is turning that off disallows use of my previous search data for model training?
Nop, he used my credentials
Somebody please help me, iam running out of money, the perplexity contact is not replying 🥺
Hey @noble spruce!
If you find the original message helpful, please consider reacting to it with the :star: emoji. If the post is appreciated by the community and receives 5 stars, it will go to the https://discord.com/channels/1047197230748151888/1082806833938436228 channel and the post author will get the <@&1082034222778302614> role on Perplexity.
dm @slim night , he might help you on this case
How can I take a picture using the Perplexity Assistant? I see a camera icon at the bottom left, but there is no shutter button. I have tried various commands like 'click the picture' but none seem to work
Hello, my trail is finished in 3 days. Perplexity is the better way I can explore 😉 But, it's too expensive for me. Non way to have a teacher plan or a offer similar at the black Friday ? Thanks for information
i have a mac and i use perplexity ai on a web browser, is there any way of perplexity ai being able to look at images and give you information about the image, for example, if i gave perplexity ai a graphical problem and i wanted it to solve the graphical problem, could it do it?
I’m new to Perplexity. I’m not a pro user yet but trying it. I use it currently on iPhone. I am low-vision so I would love to use the audio function for most things. Can I use my camera and have it help me navigate recipes, summarize text, etc? Is there a way to connect the API for other apps? Still learning and trying to find the best solution being I can only do 1 subscription. Thanks!
I'm confused, there is a direct function to add images, what do you need it to do that it's not currently?
i already saved the screenshot as a pdf
right, I assume if you open the PDF you can see the screenshot you took. But also, if you're uploading an image and it's a PDF file, then I can understand it might be confused.
i only uploaded the pdf
how do i overcome this problem?
open the pdf, take a screenshot of the image and save it as a jpg maybe
i cant select the jpeg file when i try 2 attach a file for some reason, its blanked out
what model are you using?
In case you are planning to be a pro user https://perplexity.ai/pro?referral_code=TQRFFK3J
Here is a discount
Hi, everyone I just joined this community. The company I belong wants to use Perplexity API to make new application and it has rule of using Gen-AI. But I'm not familiar with policy of Perplexity. I looked up the official site but I couldn't decide if I can use Perplexity API. So I need a help. I want to figure out the questions that I attach in img.
did sonar large chage it’s name to just SONAR?
Guys, the assistant on my Android doesn't recognise or respond to voice commands why
please send an email to api@perplexity.ai
Even o1 has disappeared, perhaps because R1 is coming?
I notice the same thing when I use my pc, with fewer and fewer models available. The android application offers a few more even though o1 has never been present in this application. Quite disappointed in the end
Since it was pretty vague, are certain functions of perplexity assistant that require a pro account or is the whole thing free?
Check if you need to hold to talk, and that you have the assistant enabled (go to the account settings)
Or you might need to enable permissions for voice for the app
or you haven't set perplexity to your AI assistant (i think by default it's stuck as google)
Aravind mentioned on X that r1 is available
Just updated the andriod app - still don't see r1
Am I supposed to see it here?
Apologies for my unawareness
its in pro search section
its in pro search section
THIS is such a scam
refund button is such a fob off its not there for anyone
does flux.1 work well for anyone for image generation?
no matter what prompt I use it's pure garbage, but when I put the same prompt into the flux.1 webserver, it works perfeclty
dall-e and stable diffusion also work fine, but I like the flux model results
perplexity must be poorly configured right?
Deepseek not working
What’s the best way to run these models locally? Just got a 32gb ram laptop with a 1TB ssd (assuming this even enough to run anything at all)
An easy way to get started with running models locally are tools like JanAI or LM Studio
what really matters is the GPU, RAM is good for holding larger models in memory, but the GPU is the one doing all the work. LM studio and GPT4all are great options that are easy to get going.
Thank you both! GPU is an rtx 3050. Gonna get started w LM studio and see how it goes
android ver doesnt have an o1 and r1 option, only on the web browser version
Where are some good prompts to be used with spaces in Perplexity
Why can’t I see Deepseek on my perplexity pro ios app?
the option is not available on android and ios yet you have to use in the web version
I am a beginner here.. Can someone please clarify why Perplexity AI responds like this? Perplexity is supposed to be giving exact answers with the right citation as per claims. Why this? Kindly clarify 🙏🏼
Why is the o1 model gone?
Not a developer of the app, also new here.
Seems to me that the model that generated the second answer is one of the OpenAI models.
My understanding is that Perplexity uses different generative models in the background to come up with your results.
Asking a model about perplexity, give it context on what you want to know, that it does not necessarily have when asking a model about 'itself'.
Please keep in mind that I use the whole anthropomorphising language for ease of understanding, not because the models are actually anthropomorphic in my mind.
Bascially, as with most systems, the 'better/more precise' question, has a higher chance of giving you 'better/higher quality' results.
I got it, Perplexity randomly assigns the question to one of the language models and that's why this answer comes up. But I have been seeing for the last 2 weeks, all the time this is the answer. In that case, why only open ai takes up the question ? Is it because Open AI has given excessive free API calls ?
Hello, pro member here, I have a quick question about Spaces and the stored file sources. I'm trying to put together a internal knowledge base that helps my team find digital assets. One of the files I have uploaded is a MS Powerpoint deck that contains example images for each file. I tried adding a custom prompt to say.
When the users asks for examples of [ASSET TYPE] it will first search this PPT file and return results from that first before any web searches.
It did this successfully for the text that appears in the PPT file but no images are pulled in, which is the main thing I was hoping it would do.
How else can I achieve the search being able to show results from a custom image library? I was thinking maybe it might need to go on Pinterest or something else that could be pulled in easier than images embedded in a ppt slide. But thought I would ask here as it wasn't clear from the Perplexity help guides.
Hey @blazing sparrow!
If you find the original message helpful, please consider reacting to it with the :star: emoji. If the post is appreciated by the community and receives 5 stars, it will go to the https://discord.com/channels/1047197230748151888/1082806833938436228 channel and the post author will get the <@&1082034222778302614> role on Perplexity.
When you ask the model who it's owner is, it will obviously give you the answer of the model. Context in prompts is key to get correct information.
Have you tried asking the AI to create the prompt for you which achieves the goal you'd like?
It can sometimes lead to much more accurate results (not always)
As in ask another model (ChatGPT) to write a Perplexity system prompt, to extract the images for the ppt file?
see if this helps
I copy pastad some of your request and just changed some wording a bit
idk if it will fix the issue or not, but sometimes these create better prompts than we make
Thanks for trying but it's still not managed to pull any images out.
dang, sorry to hear
So I’m trying to make “Custom GPT’s” in Perplexity. I am using Spaces, and it does have a section for a custom prompt. I put the exact same one I put in ChatGPT, and then saved it.
It’s a prompt that requires me to upload a document that has been transcribed, and the custom prompt already has all the instructions, so I was hoping just to upload the file and click send, but it’s asking me to type something in too. What should I do?
I can't find deepseek even in web version as well.
You should write something
My prompt has the instructions. It literally says I'm going to submit a file and then bla blah bla
So I dont really need to say anything more. The prompt is there.
For comparison, oh my custom GPT I just plop the file in a new chat and press send
ok? But it's unfortunately not how Perplexity is wanting to function, so the answer is simply write something.
Make Custom "GPTs" on every platform, and every platform wants something different.
Gemini is similar, it won't accept nothing in the text box
Hi, will DeepSeek R1 be made available to those on the android mobile app?
Do you know a decent prompt to get it to just do what the custom prompt says? Or should I be making a text expander that essentially types it in every time and forget about Spaces?
I have some custom GPTs/Gems/etc made for language learning that I have given similar instructions to (specifying when a document is uploaded to create a language lesson for it), and for those I simply write "Make a lesson from this document" in the prompt.
So whatever the result you'd like is, just specify that will be the prompt to trigger the rest of the processing.
What's the best way to know if Perplexity can access/scrape a link?
ask it to, probably
I am a new enterprise pro user. We created our first space, with several users. However we aren't able to see each others threads created in the space. We can see the shared files and links, but not threads. Any advice?
Are there any exampels of connecting Notion with Perplexity?
I am looking into the Enterprise tier as our team has a deep knowledgebase on Notion, and would like for it to become searchable with Perplexity
which version of deepseek-R1 did perplexity deploy/use ? distilled models? or the 671B model ?
Hi so deepseek model is not available on web, ios or macos for a US pro user. I went 8 rounds back on forth sharing screenshots to prove same with perplexity support (who were clueless) yet the CEO and others proclaim it's available...what am I missing? of course apps are updated etc etc.
atm web version only
it's definitely on the web
nope just checked
click on pro settings on a new thread
got it
How can i reach the support for account?
Thank you, I’ll give that a shot.
please contact support@perplexity.ai
how do i know which model i had used to generate an answer in my history of chats?
Thanks
Contact @slim night as well
Just wondering if anyone else is getting extremely short responses to their Pro searches all of the sudden? I'm used to getting lengthy and detailed responses but now I can't get more than a short paragraph, or possibly two.
How do i use a promo code if I already have an actively monthly subscription? The promo will expire soon so I'd like to add it now before my subscription expires
It's been four days, I'm didn't got any response from support team through email and intercom chat. Why? Anyone please help me
Hey @noble spruce!
If you find the original message helpful, please consider reacting to it with the :star: emoji. If the post is appreciated by the community and receives 5 stars, it will go to the https://discord.com/channels/1047197230748151888/1082806833938436228 channel and the post author will get the <@&1082034222778302614> role on Perplexity.
Hi I have a pro subscription from SingTel app but it isn’t registered and I only have 3 pro searches a day. Help
Hello, is there a limit of the output I can get if I ask a question?
@slim night
Thank you
I read an article on reddit about whether pro users are now allowed to use the r1 and o1 reasoning feature 25 times a day, not 10, is there a clear announcement about this?
The daily limit has been increased from 10 to 25 DeepSeek R1 queries a day for Perplexity Pro users. Goal is to keep increasing this as we add more capacity! Enjoy.
Hi, how can I redeem promo code ? The page showing that code applied but nothing change. I have also wrote on Monday to Customer Service but no one has replied yet
@slim night
which deepseek r1 does the mac app use? is it the full 671B?
Hi so i am on free tier (tier 0) can i not make any api calls or generate an api key?
you have to top up the api account first
Is there a way to keep Perplexity on track of the current "Code File".
I keep getting lost over long conversations and perplexity keeps grabbing old code and tries to remake it, instead taking the last current one I send.
I want to test out GPT since they give a nice way of having a "Answer Window" that you can let the AI edit instead of repasting the same code u want to edit.
What are the details on Perplexity's Deepseek instance? When I ask Perplexity, it tells me the data is stored in China and all the Deepseek T&C for that country apply. Also, i'm told the IP address resolves to servers in China. I suspect this is incorrect so I'm just looking for something written by Perplexity that says more than just it is in the US. If there is a blog post or something from the company, that is enough for me.
I am wondering the same thing. If I understand correctly, the sonar release will allow us to leverage Deepseek with out the privacy concerns of our data being stored in China? Am I correct?
The assistant on android is working pretty well for me but I was curious why it is using a male voice when I have my voice style in app set to Kate? It does give Kate when in app
1000 duck sized chickens or 100 chicken sized ducks?
Hi! How I can talk with a support? (I'm a Pro user)
i think the default in R1 is very compact responses. Ask it to ignore brevity.
The first question is about perplexity the product. The second question is about the language model user is interacting with and that depends on the specific AI. As Perplexity does not develop their own models you can’t ask that question to that type of model. HTH.
@slim night o1 is missing in the app
Is that accidental or intentional?
The DeepSeek api can answer it but only the desktop and app can’t
Hi, Are there any plans to integrate perplexity assistant with Android auto?
@ok.alex o1 is missing in the app
support@perplexity.ai or via intercom pro support -> https://www.perplexity.ai/settings/account
Hi, I am on Pro but I can’t find R1 option. Any idea?
Thanks for reporting, the team is checking.
Thanks for replying
could we expect longer reponse limit on reasoning models as the task given to them require longer answers?
There was a post about this earlier)
https://discord.com/channels/1047197230748151888/1333077849665704010
But no response.
thats sad , the thread shows # unknown to me btw
any news? It's a bit frustating can't use o1
Am I reading it right that the sonar-reasoning call include COT text garbage with it? I don't see why I would want that in all cases and not delineated as a separate object.
Hi Alex,
I’m following up on my refund request. This ₹2000 is all I had, and it’s critical for my dialysis. Please help me resolve this as soon as possible.
How can I best keep perplexity from losing previous iterated work? I.E coding iterating to troubleshoot or add new features. It loses features as we go.
I have been trying to contact someone from Perplexity to refund an erroneous charge on my CC.
How do I get in contact with someone, anyone? Phone number is a google voice account, and chat window goes unanswered for the last couple of day.
qq on perplexity's sonar api, has anyone successfully got the search_domain_filter request param working in the /chat/completions endpoint? https://docs.perplexity.ai/api-reference/chat-completions#body-search-domain-filter
Generates a model's response for the given chat conversation.
and... i just searched for search_domain_filter and found similar/same issue reported by others. reading more.
I have Perplexity Pro.
I asked it to read a PDF file (4 MB; 46 pages) and summarize job descriptions within.
The answer looked good, but was incomplete and truncated to 5,254 characters.
Q: What is the limitation that I reached? How might I get a complete answer? (Example. In my prompt, constrain the answer length?)
In the interim, I used follow up questions to split my answer into chunks.
anyone else having problems signing into perplexity? i put in my email and it sends a temp link, and either it takes ages to arrive, or never comes and times out?
NEW
Hello, new user here 👋
I created a page from a thread. I asked questions in French and got replies in French. During the process to convert to a page it did the intro in French, but the rest of the page in English. How can I fix it so that is all in French?
🔗 https://www.perplexity.ai/page/bilan-de-hilo-dhBMnjO6RJuC55MLni1AEw
I currently have a Pro account and am very interested in the enterprise account for my firm. Essentially, I have (and will continue to) upload standards and workflows to a space on perplexity. I would like to embed this space into a Microsoft Teams Tab and allow all the staff within my firm, the ability to ask it questions. Ideally, they dont have to be invited or create an account. It would be internal, but publically internall if that makes sense. Is this something I can do with the enterprise account? What additional costs or limitations should I consider before moving on this.
Was curious if Perplexity has opportunities for sponsorships? Saw that they were featured in one of my favorite Youtube channels + was wondering if there's a place to apply 👀
Perplexity remove r1 model but only keeps the r1 search
I don't see Perplexity in iOS share sheet, if I want, let's say, to forward a news article to it while I am browsing the web with Safari. Is it just me or Perplexity just forgot?
Hi, I try to use perplexity api and ask json format. But everytime it startswith triple backticks like ``json so I cannot parse json easily.
If I run doc api code.
https://docs.perplexity.ai/guides/structured-outputs
it returns like this
Can perplexity parse links that arent optimized for LLMs? for example
reading all the articles in this website
https://www.strongerbyscience.com/articles/#
Perplexity removes the o1 model and the r1 model and turns it into the r1 search and o1 search 😭😩 can I have both search and model???
What does "Uncensored" mean when it comes to AI Models? Freedom of Speech? NSFW? Just curious, still learning.
Based on this
The app and website censored because of government except the api DeepSeek
Sonar reasoning is the less censored but still censored version but the api is very uncensored api DeepSeek
what are the diffrences between pro and enterprise
Pro for individual while enterprise for the company
i mean yeah... but the price is almost double? there has to be some huge diffrence
Enterprise for a lot of people uses one account while the pro uses one person one account
In your opinions, which have you found to be best, Perplexity o1, R1, or Pro search?
Why we have two place to select the R1
AI model and something else in the text input bar
Hello, I remember that we could select o1 before, this doesn't seem to be possible anymore. Am I missing something?
They remove it
O1 turn to r1 when using the desktop version
Thank you
It's a reasonable decision, given the superiority of the R1.
O1 is smarter than r1 and r1 is less censored than o1
The DeepSeek very uncensored API not the website or app
DeepSeek website is run from China and censored under Chinese law. Perplexity runs their own DeepSeek instance(s) and does not censor output, per public comments from Perplexity CEO, Srinivas:
https://opentools.ai/news/perplexity-ais-bold-move-integrating-deepseek-r1-to-bypass-censorship
Hi, all! New Perplexity Pro subscriber, long-time user.
Question: is there a way to export my usage statistics and/or query history? Am curious to see just how often I've used the free restults before finally subscribing.
Hi !
I'm moving to Perplexity from Claude. I'm unable to make the native voice search work similar to the Claude app. Is there any settings I need to enable ?
Note : Not the assistant. Just the in app voice search.
Hello! Any papers or code available on how Perplexity does RAG and sources under the hood?
Links to technical reports or blog posts also useful!
I'm suddenly using deepseek out of the blue??? I went to options and a) I don't see deepseek as an option and b) it shows I still have GPT-4o selected (web app)
it looks like sonar is deepseek now? Still I don't have it selected, also I'm not on mac
I like the idea of it but it is SLOOOOOW
Ok, I actually found the toggle for it, it's next to the pro selector, but I don't have either reasoning enabled
Hi everyone. My Mac desktop app defaults to the R1 model even when I've selected the Default reasoning model. When I toggle the models to Claude or GPT, it still defaults to R1. Anyone else facing the same issue?
yea that's my problem but on web. I think they just pushed an update that fixes it on web though
Do you even test the API???
Deepseek reasoning API is very uncensored but the app and website are censored by law
Sonar reasoning API the us version of Deepseek is not completely uncensored
Less strict but still censored
Perplexity removed the r1 model yesterday
censored by law? what law, isn't that exactly what freedom of speech is for? to prevent the government from censoring
also it's back, if they did remove it
otherwise we wouldn't be here complaining that it's enabled by default and forced
Still missing
Perplexity remove it yesterday
The model not the search
what is that line that says pro reasoning with R1
The Deepseek API didn't censor the truth but only the app and website because china law
If you want to know the truth use the Deepseek API after they did their maintenance
OH you meant the official deepseek, I thought you were saying the one here (which they explicitly say is uncensored)
I said both Deepseek api very uncensored while the Deepseek us version api still censored but less strict
You show search while they got both r1 search and r1 model but sadly removed the r1 model but kept the r1 search
Now we only have less intelligence model
I've been using Perplexity R1 all evening.
I see. Great, thanks for the information. I'm still getting the same issue on Mac but on iOS it works fine. Will try out the web version.
You use the search r1 all evening
Both are available right now.
I already, did
Where is the model?
I just did. Look at the image.
You show search but I show the model
What is the difference?
Doesn't search just use whatever model you choose
It was yesterday perplexity remove the model version of r1
We got less intelligence model using search r1 but it’s better get intelligence model with r1 o3 search for better result
It is still there right now, they just remove the little blue tag that says new.
I actually don't have it on web, even though I still have the new flag 
Maybe it's only activated on some accounts?
actually, do you not have grok-2?
I do have grok-2, I just didn't include the bottom of the screenshot because that's not what we're talking about.
I was using it on web all evening, I'm only giving screenshots of the app on Android because I'm now outside walking my dogs.
Does anyone know the answer to this?
I feel like an idiot, but did they remove Grok and Sonar from the rewrite section?
i have perplexity pro and rejoined the server but still havent received access to the pro channels
I think it's just a voice channel, but idk why we don't get the tag
I don't but that's a really good question, I'm interested too
dang, well i hope they add it sometime
They keep for rewrite with r1 but not the model
Now no grok 2 and no sonar
@slim night So many good models are missing now
Still no r1 even I sign in and sign out and sign in and even use desktop version of the website
why does mobile have more models than website?
I didn't find any way to download all data, but the threads list should include all your primary (first) searches that weren't incognito or deleted. Annoyingly they don't all load at once, they live load... maybe the full list is in the html?
The mobile doesn't have o3 search
neither does website apparently
unless its 4o
in which case then i'm stupid lol
I'm not gonna make any direct suggestions, but take a look at #2. https://www.perplexity.ai/search/is-there-a-way-to-download-my-24Z1Ai9PRZG8RECVABaW0A
I guess tomorrow I can go look at the html.
The mention of a Chrome extension excited me at first, but it only allows you to save the specific chat you're in, so it would still require going through all of them.
ah rip, I didn't full look at it 😦
is there a way i can gift to my brother perplexity pro?
what is wrong or the drawback of the following thinking: because perplexity has access to multiple llms, and especially now that it has two reasoning llms, it would be a good idea to attempt to get the llms to cooperate to provide a better answer
It’s better getting smarter model
Perplexity keep removing the good model
I noticed that, in my PC, when I open a previous discussion with Perplexity, there is an arrow that takes me to the bottom of the chat (to the most recent messages). I did not notice this arrow in the phone app. Is there any way to go fast to the bottom of the chat (to the most recent messages) in the phone app? (maybe the arrow is there and I'm not noticing it?!)
why does perplexity have 0 memory, I've been working on a offer deck and I gave perplexity a simple rule: write in a pragraph as if it was a conversation and I had to do that every time even though I said to do that every time.
Hi, do you know if the new O3-mini option in Perplexity is better than the 01 model that was before and what about Deepseek R1?
i think it's a bit dumb compared to chatgpt version
Anyone still having issues with the Mac app defaulting to using R1, regardless of which AI Model is selected under Settings?
I’m on iOS but same thing
Hey, the maximum number of questions has changed a few times. How many times a day can I make prompts with o3-mini?
500
What do you say good people?
Can i set a personalized global prompt so that LLMs can use for all my future prompts?
Input Custom Instructions for the AI. (Space) https://www.perplexity.ai/hub/faq/what-are-spaces
You can also use your profile to a lesser extent: https://www.perplexity.ai/settings/profile
what are the usage limits for chatting with different models when I am not using the search?
Why did they change from o1 to o3 mini
good question, i think because it's cheaper
It’s newer
Is anyone still having the problem with the R1 selecting even though you selected something else.
And confusing since perplexity keeps changing everyday
Yeah
now i can't use both r1 and o3, has anyone the same issue?
really really bad days for perplexity users
Yes I’m so mad
Why Can it be default model for o3 mini and R1?
Why must be turn on pro button to get the o3 mini and r1?
So frustrating on why we get less intelligence models now?
You ever get help? Currently encountering same issues and can’t get response on refund
i am currently using perplexity pro to help me with some coding stuff and i am running the cluad model however deepseek keeps being used does anyone know how to turn this off
yes I am and its pissing me off I dont want deepseek i want to use cluad
Is anyone else have that problem? the mobile app keep using deepseek even when i tell it to use preplexity pro
its doing it on both the meb version and the desktop veersion
the problem started for me late last night like 12 am
I didnt had problem on desktop
And i was using that just few hours ago its the mobile that cuaae prploblem for me i hope they will fix it soon becuase this is some series bug
It’s making me upset
i am pretty sure its still using the R1 since its still giving me the same bad answers as before respoonse that cluad does not do
i am actually in the porcess of talking with support
Have they said anything
Im joining the question
Nobody has o3
How do I turn r1 off?!
Idk but it’s making me maddd
Yeah
Same question here
