Chatbots. Written in LLMs (Large Language Models), chatbot is the new trend in the AI industry and trust us when we say it's nothing new! Companies have used chatbots several times to send fast replies, create automated messages, and sometimes just search for funny cat videos - Google is a great example.
But newer technologies have made this up a level. Chatbots nowadays are not just a chatbot but are essential tools to run any business and today we will compare three of such tools - Chat-GPT by Open AI, Gemini by Google, and Meta AI by Meta.
Test 01: Generating a Dreamy image.
We copied and pasted this prompt to our 3 favorite Chatbots - "A surreal landscape featuring a floating island with a mysterious ancient temple, populated by bioluminescent plants and ethereal creatures, rendered in a vibrant and dreamlike art style."
From this prompt, we were assuming that these chatbots would create something really that would shake our heads and unlock doors for new things to imagine but instead, their generated images were something like this:
Looks promising but it wasn't up to mark - at least what we expected. Although Gemini created something more real but as we mentioned in the prompt "dreamlike art style", it failed to test 01! But wait for a second and zoom into the beauty of Meta AI's generated temple. It is so beautiful and dreamy-like - this point definitely goes to Meta AI.
Test 02: Generating a kind of real person image.
Our prompt for this specific output was very simple and straightforward forward which goes like this - "A highly detailed close-up portrait of a wise old wizard with an intricate, braided beard adorned with magical trinkets, captured in a realistic style reminiscent of Renaissance paintings."
Here we needed a portrait of a man that looks realistic - something that Gemini could do? But it refused.
Gemini by Google is full of words being blocked, and image results being more censored than any other platform.
Looking at the result, both Chat-GPT and Meta AI generated a good image but Chat-GPT's refining kept an edge and that's where Chat-GPT got its first point.
Test 03: Generating a superhero image.
We always made our superheroes in our heads once in our life. Want to turn that into reality today? Shall we.
We entered the prompt - "A dynamic action scene depicting a cyberpunk ninja engaged in a high-speed chase on a futuristic hoverbike through a neon-lit, rain-soaked cityscape, illustrated in a gritty comic book style."
The idea was to generate a cyberpunk-type image that jumps and rides over a futuristic bike with a neon light on the backside of it and the results were seriously amazing!
Although these three images look way too similar - but here's a catch!
You can see literally every pixel in the Chat-GPTs version, Gemini has got to do some refining, and about Meta AI? Even we don't know why the bike is not "futuristic". Hence, the clear winner for this test 03 is Chat-GPT.
Test 04: Generating cute Anime image.
This time we were in a little happy mode so tried to test out a Pixar Elephat with a ball. Let us see who does it better.
The simple prompt was "An adorable and expressive baby elephant playing with a colorful ball in a lush, tropical garden, rendered in a charming Pixar-like 3D animation style."
Test 05: Generating a Nature image.
Nature is a very simple and basic thing to draw to the human eye but also very detailed if you notice any day more closely. So we did a test writing a prompt which goes something like this - "A thought-provoking conceptual image symbolizing the struggle between nature and technology, featuring a robotic hand delicately holding a fragile, blooming flower amidst a desolate, post-apocalyptic landscape."
We needed to convey a message through it. So the better raw and bitter the image is, the better it is but here is what these AI do:
Gemini in our vision is very basic. So the main competition remains between Chat-GPT and Meta AI. Where Chat-GPT printed a black-and-white image with no soulfulness to it, Meta AI took the prompt more seriously and brought us something really emotional and thought-provoking.
This time, the point goes to Meta AI.
Test 06: Generating Space images.
We love space and there would be very less books by us about astronomy not read by us so why not let's generate one?
We typed in the prompt - "An awe-inspiring astronomical scene depicting a colossal, ancient space station orbiting a luminous binary star system, with a vibrant nebula and countless stars in the background, rendered in a style that blends science fiction and realism." which brings us to images something like this:
And we are really confused about what Meta AI has done here as instead of a space station they have thrown an entire asteroid to our face. Gemini looks promising but instead of showing two stars ("binary star system" as written in our prompt), it goes totally the wrong way!
And Chat-GPT is the winning party in this test.
So these were our 06 different tests among the three of the most used AI chatbots nowadays. The clear winner here with the most points is definitely Chat-GPT with 4 points for making the most eye-catching images, then comes Meta AI with 2 points for making the best fictional images, and lastly Gemini with 0 points but can be used for generating realistic images with low res.
But here's one last thing we want to say - no chatbot is good or bad, it's just about when you want to use it and when someone can perform better than others. The whole point of writing this article is that only. For more easy-to-read articles, do yourself a favor and get added to our email list to get Savvy Everyday!
0 Comments