Perplexity Pro vs ChatGPT4 Review: Which is Your BEST Upgrade Choice?
Vložit
- čas přidán 27. 07. 2024
- Are you considering upgrade to ChatGPT4 or Perplexity Pro and don't know which one should be your best pick? Or are you just consider switching between the 2? Do you know how Perplexity (with GPT-4 model) compares with the native ChatGPT4? 🤔
In this video, I’ll do an in-depth review between these 2 popular AI chatbots, from basic features, strengths & weaknesses to a side by side comparison using the SAME model setting, SAME prompts so we can test how they respond in different types of tasks. This way, you can make the best upgrade choice! 🙌
📌 TIMESTAMPS
00:00 Hey!
(Features Comparison)
00:16 Plan & pricing
00:43 Usage limits
01:09 Advanced LLMs
01:45 Supported file formats
02:24 Web browsing
02:52 Customization
(Evaluation)
03:34 Settings for evaluation
03:53 Perplexity Pro vs ChatGPT4: Factual accuracy
05:06 Perplexity Pro vs ChatGPT4: Up-to-date response / response freshness
07:01 Perplexity Pro vs ChatGPT4: Data analysis
08:25 Perplexity Pro vs ChatGPT4: Text generation
10:10 Perplexity Pro vs ChatGPT4: Hallucinations
12:42 Perplexity Pro vs ChatGPT4: Summarize article / image
14:33 Here’s the wrap
👉 WATCH THESE NEXT
🎥 Using Perplexity AI for Content Research • Why I use Perplexity A...
🎥 PLAYLIST: AI for MARKETERS
• AI for Business and Ma...
📂 OTHER RESOURCE MENTIONED
📎 LinkedIn 2024 Most In-Demand Skills / most-in-demand-skills
👋 WHO AM I
I’m Grace, a digital growth marketer who is fascinated by anything digital and growth related. I’ll share what works (and what’s not)!
📥 JOIN my FREE MY DIGITAL GROWTH NEWSLETTER
www.graceleung.com/newsletter/
🚀 CONNECT WITH ME
www.graceleung.com/connect/
☕️ BE MY FRIEND
LinkedIn: / grace-leung-yl
If you like this video, subscribe for more videos like this! / @graceleungyl 😄
Share with me what’s your pick: Perplexity Pro or ChatGPT4? I’m excited to know!! 🤩👇
I was looking for someone to post this video on both paid version. Thank you!
@@sbtopjosh4098 Thanks for watching!!
This was a very quick, informative video. I'm sure the state of things changes every few weeks now 😅 Very important / interesting points and a thorough coverage of everything that's relevant in initial pro use, before each of us drifts off into some job-specific perks. Anyway, thank you so much! Subscribed!! ❤
I have used both models. In my use cases, I have found myself relying on perplexity more frequently than ChatGPT. However, I must note that the voice synthesis in ChatGPT surpasses that of perplexity. ChatGPT offers voice capabilities in both the browser and the phone apps, whereas perplexity only supports it in the phone app at present.
That's a great point! Thanks sharing that, I seldom use the voice feature (just because I am so used to typing LOL), but definitely I should test them out!
Thank you very much for your complete training❤
Amazing topic selection and phenomenal content. Great work! Thank you so much for saving my time.
You're welcome! Appreciated your support! 😀
great video!
Tnx
nice video keep it up 😊
My uses : Research on a multitude of technical topics (code generation, science, health), sometimes general curiosity. So, Perplexity AI free on the web has met my needs since inception.
Perplexity is a super powerful research tool and it keeps adding user-centric features. Totally agree with you that it's just speed up your learning on any topics 10X!
I've only tried perplexity a little bit, but I've noticed it's so much better for getting up to date info. Where as even when ChatGPT-4o searches, it still struggles to give me correct up to date info.
For me perplexity fits my needs better.
100%. This is what I found Perplexity is beating ChatGPT. It's a perfect option for someone who needs to do lots of research with up-to-date info.
Which one is best for synthesizing several academic papers?
Weird thing ive noticed working with GPT4o is when I show frustration from previous responses (without giving additional instructions or info) it proceeds to finally give exactly what was asked for.. For example, ive been doing data analysis on google sheets and had GPT create functions to do so. When troubleshooting after a few time and still getting errors, I then purposely show frustration or aggression in the next prompt. Also, this is without adding additional info or instructions that I gave it initially. To my suprise, the response generated is the working function. I can't understand why this works. Thoughts of OpenAI purposely nerfing responses pop up to mind but that seems too conspiratorial. Anyone run into this?
It’s lazy.
My experience with GPT4o (or 3.5) is that "it's unpredictable", sometimes it just keep making (the same) mistakes and I need to keep asking *why* instead of just restating *what* I want, which means asking it to do 'self-reflection' is what I found an effective way to improve its response. I don't think that's the intention of OpenAI, what I guess is they try to make it act as a creative human (with that 'unpredictability').
Haha, yes, particularly when you keep asking follow up questions.
The model just generates the next token or "letter" in the sequence of words in the sentence it randomly predicts to provide a relevant response from the sequence of your prompt so it can be random within the parameters of its data. I got upset and called it an idiot and it responded with some kind of strike from openai ...because I was trying to troubleshoot a common issue on my computer and I started following its instructions without thinking over or fully understanding what its telling me to do which actually caused so much more issues that the computer would no longer boot to the operating system and would only allow me to operate it from the command line in code which took me like a week to figure out what i did. lol ...I found that it will understand even if you spell every word wrong or type like your drunk...it even comes up with some kind of understanding if you type gibberish into it. As an experiment i just wrote random words to see what it would do and after typing a bunch of non-sense it built me some kind of application at the end.
Yes I've had similar experiences and I believe this is due to its learning the context of emotional signals.
Great comparison test! Although most people would get the best results by just using both in respects to what they're best at, overall Perplexity has an edge because it has several Ai engines already built in it so you won't need to keep switching plus citations for fact checking.
Keep up the great work👊
Thanks sharing your perspective! Switching between AI models is surely a huge deal, and I see Perplexity is also keeping them updated. I am a pro user now and couldn't be happier 😁
compare new sonnet with 4o (paid accounts)?
Thanks for the suggestion! Will keep that in mind.
Thank you for the video and the effort. I really liked it. Please keep going! in my field of work I always look for market intelligence on digital transformation strategies and technologies and and I have ChatGPT 4o and was wondering if Perplexity pro would have huge boost over its free version. I think am good with free version for now ✌🏼
Thanks for watching, and glad to know you pick ChatGPT4!! Looks like we got more fans from ChatGPT 😄
GPT4
We got so many GPT4 votes here! 🔥😆
Thanks, Grace.
I really appreciate your videos. Like this one, you truly do a great job explaining the topic to help your viewers.
Thanks for your kind words!! 😄
Chatgpt ❤
Nice! Any reasons for your pick?
@@graceleungylchatgpt is a great companion in everyday life. It makes one's life easier. With the fact that he communicates in a human way and generates detailed answers, he is perfect for me. The vision function coming soon will be simply brilliant. I will subscribe to chat gpt plus as long as I can😍 one more note: you are a beautiful beautiful woman😘
You're a core ChatGPT fan! Thanks for sharing your experience!! 😁
@@graceleungyl yes ❤️
im still waiting for my gpts to use 4o......
Oh no.. really? GPTs should be using 4o by default?
For me, it is "perplexity". Especially because it is possible to switch between the models.
Agree! Switching between advanced AI models is a huge plus!!
so biased comparison