Just when you started coming to terms with ChatGPT's eerie capabilities,Swipe OpenAI dropped a new version of its AI language model.
OpenAI says GPT-4 is much more advanced than GPT-3, which powers ChatGPT. And to prove it, they made GPT-4 sit down for a bunch of exams. OpenAI tested GPT-4 with a variety of standardized tests from high school to graduate to professional level and spanning across mathematics, science, coding, history, literature, and even the one you take to become a sommelier. The exams were comprised of multiple choice and free-response question and GPT-4 was scored using the standard methodology for each exam.
SEE ALSO: How to get access to GPT-4 right nowPut your pencil down, GPT-4, it's time to see check your scores.
GPT-4 didn't just get into law school, it passed the bar. The AI language model scored in the 88th percentile on the LSATs (Law School Admission Test) and did even better on the Bar (Uniform Bar Exam) by scoring in the 90th percentile. By comparison, GPT-3 was in the bottom 40 percent of the LSATs and 10 percent on the Bar.
GPT-4 took both the math and reading/writing sections of the SATs and all three sections of the GREs which are broken down into quantitative, verbal, and writing skills. It scored in the 80th or 90th percentile of all sections except for the writing section of the GREs... which it kind of bombed in the 54th percentile.
The quintessential overachiever, GPT-4 also took allthe AP (Advanced Placement) high school exams. It aced most of them, scoring between the 84th and 100th, except for a few outliers.
GPT-4 scored 44th in AP English Language and a measly 22nd in AP English Literature. So all you wordsmiths out there might have some more time before GPT-4 replaces you. GPT-4 didn't do so hot on AP Calculus BC scoring between 43rd and 59th, proving that even for a supercomputer, calculus is not easy. But that still earns GPT-4 a four, so it might still place out of college calculus.
GPT-4 still has some work to do with its coding skills, which is curious since one of its marketed uses is for helping developers. Its rating for Codeforces, which hosts competitive programming events, is 392, which puts it way down in the Newbie category of anything below 1199.
It did pretty well on the easy level of the Leetcode (31 out of 41 problems solved) but struggled when it came to medium or hard level of difficulty (21/80 and 3/45 respectively). As we saw in the developer demo livestream, GPT-4 is fully capable of writing Python, but required some manual tweaking to set the right parameters, which might explain some these test scores. Or maybe it didn't eat breakfast that morning.
GPT-4 passed the sommelier exams with flying colors. It placed lowest (77th percentile) in the most advanced sommelier exam. But for a non-human entity that's never tasted wine, we'll let that one slide.
OpenAI has released a full breakdown of how GPT-4 performed. GPT-4 might not write the next great American novel...yet, but GPT-4's future as a mathematically brilliant lawyer and wine connoisseur looks pretty bright.
Topics Artificial Intelligence ChatGPT
Whale Vomit Episode 5: Startup Monarchy
'Love is Blind''s live reunion crashed Netflix. The internet reacted.
Writers’ Fridges: Carmen Maria Machado
Early Cyber Monday: Lego Ideas BTS Dynamite kit on sale for $64.99 at Amazon
Fyre Festival and Trump’s Language
Who Gets to Be a Mad Scientist?
America’s First Female Mapmaker by Ted Widmer
Ode to the Dinkus
Ireland fines TikTok $600 million for sharing user data with China
15+ early Cyber Monday 2023 Chromebook deals at Best Buy
How to Easily Make iPhone Ringtones Using Only iTunes
Poetry Rx: Won’t You Celebrate with Me?
What's Queer Form Anyway? An Interview with Maggie Nelson
YouTubers Jacksepticeye and CrankGameplays launch new podcast 'Brain Leak'
Shop Owala's Memorial Day Sale for 30% off tumblers
Best early Cyber Monday vacuum deals 2023: Shark, Dyson, and more
The Man Behind the Weegee
'Love is Blind''s live reunion crashed Netflix. The internet reacted.
Best robot vacuum deal: Save $200 on Eufy X10 Pro Omni robot vacuum
50+ best early Cyber Monday monitor deals: Save up to $800
9 gifts teachers really want for Teacher Appreciation DayWhat to expect from '13 Reasons Why' Season 2Microsoft goes full Minority Report with Gesture API for Windows 10'Irreplaceable' plant specimens destroyed by customs officials with no chillFacebook empowers Page owners to politicize their posts'American Idol' is officially coming back because time is a flat circleChill teen shows up to prom in a hearse and casketSeduce people using this ‘slow’ dating app — if you have timeTesla's latest Autopilot update records the road while you driveNot even 'American Idol' fans want 'American Idol' to come backSouth Korea is building the world's biggest selfScared, lonely and confused: What concussions inflict upon NFL legendsThanks to Amazon, it's time to kill your landlineSenator ripped for not understanding the meaning of 'unclassified'Facebook is going to do something about those terrible ads on your websitePeople are cutting up Ikea tote bags to make weird and wonderful creationsWindows 10 hits 500 million usersChill teen shows up to prom in a hearse and casketSnapchat quietly released new geofilters for nearby businessesMicrosoft CEO: It's our job to prevent '1984' from coming true Uber to charge some drivers for the chance to earn more Timothy Simons and Tony Hale make video for Julia Louis Twitter outraged for Janet Jackson after Justin Timberlake announced for Super Bowl halftime show Bitcoin hits another milestone as it climbs past $6,000 Taylor Swift just dropped a new song and it's 'Gorgeous' Nivea's controversial skin How one resilient teen who stutters uses his art to communicate Creepy Halloween bento boxes might be too delightful to eat Hackers can do all kinds of awful things with these child smartwatches Someone built a touch Boeing gets serious about self Arianna Huffington's new Samsung app mutes notifications There's a sexy fidget spinner costume because 2017 is trash Essential Phone price slashed to $499 #MeToo hashtag has spread to #YoTambien, أنا MasterCard gets rid of signing credit card receipts The Weather Channel's Puerto Rico homepage is entirely necessary Experts don't know if fake news is going to get more or less awful Google led a $1 billion investment in Lyft, here's why that matters Glass bridge ups the ante with Instagram
2.1017s , 8223.4453125 kb
Copyright © 2025 Powered by 【Swipe】,Unobstructed Information Network