AI Models (o1, Claude, Gemini, DeepSeek, QwQ) Try a College-Level Astrophysics Problem (Highlights)

  Рет қаралды 2,098

Kyle Kabasares

Kyle Kabasares

Күн бұрын

Пікірлер: 20
@JamesHui-vl8zo
@JamesHui-vl8zo Күн бұрын
strange video idea: what you had someone generate responses for a question on each of the models, unformat everything & try to guess which model generated what?
@almatsumalmaadi8103
@almatsumalmaadi8103 2 күн бұрын
You forgot to try the new DeepSeek V3
@eigenvector123
@eigenvector123 2 күн бұрын
He did it in the previous live stream.
@wyy-xd4uj
@wyy-xd4uj Күн бұрын
It wasn't good at math.
@ashleigh3021
@ashleigh3021 19 сағат бұрын
It’s not that good. A lot of Chinese propaganda going around
@lyellbrown9766
@lyellbrown9766 Күн бұрын
Highlighting what you are doing is better than playing it at 1.5x, thumbs up.
@futurama45345
@futurama45345 2 күн бұрын
Thank you Kyle for your videos. I do suggest maybe an excel sheet with a table to keep track, which models got which qtn right or wrong. Thanks a lot
@martini1179
@martini1179 2 күн бұрын
One thing I would like to see is the AI models attempting to solve the problem live. Like you, I like watching them think.
@jamespat7975
@jamespat7975 2 күн бұрын
8 students(namely 1 to 8 ) are arrange to sit at the seats in two row of 5(meaning there are always two empty seats). Find possible numbers of arrangements in which students 1,2,3 sit next to each other, 7 and 8 not sit to each other and that two empty seats must be next to each other. My mannual hand calculations answer = 15,264, anyone here can help to use ChatGPT o1 Pro to calculate it ?
@MeridianMindset
@MeridianMindset Күн бұрын
15264 is what o1 pro give me.
@jamespat7975
@jamespat7975 Күн бұрын
@@MeridianMindset Could you please post o1 pro results here ? I would like to see the step by step calculations. Thanks in advance
@mash-room
@mash-room Күн бұрын
can you try sonus pro with reasoning? this is a new reasoning model
@maderri2
@maderri2 2 күн бұрын
So which model was the best in your opinion? O1 ?
@SimonNgai-d3u
@SimonNgai-d3u 2 күн бұрын
From your test, we can pretty much confirm o1 has proven its ability to solve novel problem just like how students taking an exam, instead of just memorizing the answers from training data!
@parthasarathyvenkatadri
@parthasarathyvenkatadri 2 күн бұрын
Asked it a simple question .... There are 100 people in a room and 90 of them know language 1 75 people know language 2 80 people know language 3 ... What could be the maximum number of people who could know all 3 languages . And all of them are hell bent on gas lighting me to say the ans is either 75 or 72 .... The ans is neither .. if you do some logic ..
@PLAY-bv1fc
@PLAY-bv1fc 2 күн бұрын
Um
@ellielikesmath
@ellielikesmath 2 күн бұрын
nope, the answer is 75 lol
@jacobshank7336
@jacobshank7336 Күн бұрын
It is indeed 75.
@katarinagorse6668
@katarinagorse6668 Күн бұрын
@@ellielikesmath No, it's not, you sillies, think about it. If 75 people know all three languages, than that leaves only 5 (remaining from the 80) + 10 (remaining from the 90) = 15 languages left to be distributed over 25 people. That would mean that at least 25 - 15 = 10 people would not know ANY language which is absurd!
@DawsonPiano
@DawsonPiano 7 сағат бұрын
so then the ans is 65?
Wednesday VS Enid: Who is The Best Mommy? #shorts
0:14
Troom Oki Toki
Рет қаралды 50 МЛН
Вопрос Ребром - Джиган
43:52
Gazgolder
Рет қаралды 3,8 МЛН
Visualizing transformers and attention | Talk for TNG Big Tech Day '24
57:45
Anthropic MCP with Ollama, No Claude? Watch This!
29:55
Chris Hay
Рет қаралды 16 М.
Is OpenAI's o1 model a breakthrough or a bust?
7:32
Steve (Builder.io)
Рет қаралды 14 М.
Gemini 2.0 Flash Tested - Is AI Better Than Humans?
8:43
In Depth Tech Reviews
Рет қаралды 47 М.
What are AI Agents?
12:29
IBM Technology
Рет қаралды 1 МЛН
DeepSeek V3 is *SHOCKINGLY* good for an OPEN SOURCE AI Model
31:55