Alex Banks’ Post

Alex Banks

Building a better future with AI

4mo

I'm amazed by this. Llama 3 on Groq makes GPT-4 look like a grandpa: I asked both models to list all the prime numbers from 1 to 1000. Llama 3 hit over 830 tokens per second(!) Plus, it generated the entire sequence again whilst GPT-4 was still inferencing. In an arena as competitive as LLMs, speeds like this really set you apart and open up hundreds of use cases. You must give it a go. If you enjoy insights like this, follow me Alex Banks for more on AI.

73 Comments

Christopher Larkin

Christopher Larkin

Head of Product/Chief Technology Officer @ Concord Technologies | Delivering Healthcare Solutions

4mo

What’s wrong with being a grandpa?

CSE | IIIT Dharwad

4mo

Groq is insanely fast, but I wonder chatGPT being slow because of too many users connected to the server at a time. It is a reason ?

Training managers on the playbook for leading high-performance teams. Entrepreneur | Writer | Advisor | Speaker | Coach | Community of 200K+ leaders.

4mo

While it’s not search, Google trained us to expect near instantaneous responses. Speed will become a competitive advantage I suspect.

Mohammed Lubbad 🍉

Mohammed Lubbad 🍉

Senior Data Scientist | IBM Certified Data Scientist | AI Researcher | Chief Technology Officer | Deep Learning & Machine Learning Expert | Public Speaker | Help businesses cut off costs up to 50%

4mo

That's impressive Llama 3 seems like a speed demon in generating prime numbers. The competition in LLMs must be intense Will definitely follow Alex Banks for more AI insights. Alex Banks

Free AI Tools & ChatGPT Prompts 🔥

Free AI Tools & ChatGPT Prompts 🔥

4mo

Impressive speed and efficiency with Llama 3, truly groundbreaking Alex Banks

UK/Japan Based Business Investor | Author of Unleashing Brilliance™ | Mentoring Owners to Business Growth/Exit | Business Turnaround Mentor

4mo

It's like lightning, and groq is awesome.

Lindsay Rosenthal

Lindsay Rosenthal

1 call to get 1 month of video | Learn about Gen Z + AI | AI Speaker

4mo

Insanely powerful!! AI is improving like crazy.

Helping Leaders Ignite Change Through Inner Growth 🌱

4mo

That's impressive speed from Llama 3 compared to GPT-4! AI advancements are fascinating. Alex Banks

👋 Johannes Hötter

👋 Johannes Hötter

Data-centric GenAI | Co-Founder Kern AI

4mo

I agree that Groq is mind-blowing. But is this comparison fair? ChatGPT started the analyzing mode, which typically means that it starts writing down the content in Python code. So even though it is hidden to the user (you can click on the "Analyzing" widget to open the code being generated, where you see that it is actually still okay-ish in speed), it still generates the prime numbers. Again, I totally agree, Groq is way faster with Llama and all, but i think in this comparison, it's a bit "unfair" since ChatGPT is generating the prime numbers twice: once in the Python code, and once in the answer.

Alexander Kolpakov

Alexander Kolpakov

4mo

This is not what an LLM is supposed to do. Neither it is possible to infer prime numbers by learning any initial segment thereof.

See more comments

To view or add a comment, sign in

More Relevant Posts

Andy Savage

Principal Software Engineer - [Golang, Linux, AWS, Cybersecurity, VPN Tech, Git IaC, Software Development, Backend Architect]
4mo
Report this post
Give Groq a try, it's a free OpenAI compatible API provider hosting the llama3 meta models. The thing is, they design their own GPUs and they completely trounce Nvidia GPUs in speed. If that doesn't mean much, just sign up and give it a try, use the same query in GPT-4 and Groq Llama3 70B and see the difference in speed. It's amazing to see results come back instantly, compared to line by line.

Alex Banks Alex Banks is an Influencer

Building a better future with AI
4mo

I'm amazed by this. Llama 3 on Groq makes GPT-4 look like a grandpa: I asked both models to list all the prime numbers from 1 to 1000. Llama 3 hit over 830 tokens per second(!) Plus, it generated the entire sequence again whilst GPT-4 was still inferencing. In an arena as competitive as LLMs, speeds like this really set you apart and open up hundreds of use cases. You must give it a go. If you enjoy insights like this, follow me Alex Banks for more on AI.
Like Comment
To view or add a comment, sign in
shahul hameed

Artificial Intelligence, Data Science Enthusiast
4mo
Report this post
What an amazing advancement. the combination of Langchain and Groq can provide a significant boost in enterprise applications. Langchain's natural language processing capabilities combined with Groq's high-performance computing can lead to remarkable advancements in various fields, from data analysis to machine learning.

Alex Banks Alex Banks is an Influencer

Building a better future with AI
4mo

I'm amazed by this. Llama 3 on Groq makes GPT-4 look like a grandpa: I asked both models to list all the prime numbers from 1 to 1000. Llama 3 hit over 830 tokens per second(!) Plus, it generated the entire sequence again whilst GPT-4 was still inferencing. In an arena as competitive as LLMs, speeds like this really set you apart and open up hundreds of use cases. You must give it a go. If you enjoy insights like this, follow me Alex Banks for more on AI.
Like Comment
To view or add a comment, sign in
Andrew Sheves

Building Proactive, AntiFragile Organizations | Carpe tomorrow
5mo
Report this post
GPT 5 might be imminent, but the bar is much higher than it was when 3.5 & 4 rolled out. I've been using Anthropic's Claude 3 family and am really impressed: Haiku is fast and neat... Sonnect writes beautifully and replaced one of the OpenAI models in DCDR... and Opus? The simulations I just ran with Opus are blowing my mind. If you are still telling yourself that you are a domain expert and that AI can't do what you can do, you might be about to get a rude surprise. I still think that AI+XP 💁♂️+🤖 is the winning combination so find out how to 'partner' with AI.
Like Comment
To view or add a comment, sign in
Gupteswar Rao

FOUNDER - School of AI
4mo
Report this post
AN ENJOYABLE GAME - Here is the link to a word game that uses AI powered by Machine Learning- https://lnkd.in/eErddqyG Play around with different types of clues to see what references the AI understands best....
Like Comment
To view or add a comment, sign in
Ethan Mollick Ethan Mollick is an Influencer
9mo
Report this post
Google’s new Gemini AI model is out. The big deal is that Gemino Ultra appears to be the first model to beat GPT-4. The fascinating thing is that it does it by just a tiny bit. For right now, they are only integrating Gemini Pro into Bard. Gemini Pro is a GPT-3.5 class model, so it will not be up to the standards of GPT-4, yet.
104 Comments
Like Comment
To view or add a comment, sign in
Mark Edmondson

Founder at Sunholo GenAI DevOps. Cloud Engineer at 8-bit-sheep/amass Google Developer Expert for Google Cloud
9mo Edited
Report this post
Initial reports is that Gemini Ultra is a GPT-4 beater, Gemini Pro is a GPT-3.5 beater. Honestly just being comparable will be huge since that means we can keep all data within GCP and not send to a 3rd party.
Ethan Mollick Ethan Mollick is an Influencer
9mo

Google’s new Gemini AI model is out. The big deal is that Gemino Ultra appears to be the first model to beat GPT-4. The fascinating thing is that it does it by just a tiny bit. For right now, they are only integrating Gemini Pro into Bard. Gemini Pro is a GPT-3.5 class model, so it will not be up to the standards of GPT-4, yet.
9 Comments
Like Comment
To view or add a comment, sign in
Harvey Castro, MD, MBA. Harvey Castro, MD, MBA. is an Influencer

AI in Healthcare Expert, Keynote Speaker, GPT Advisor, Former-CEO Of A Healthcare company, Author. Leading the AI healthcare revolution, developing & advising on AI applications. Recovering ER doctor turned AI futurist.
9mo
Report this post
I will be writing more on this new ai!
Ethan Mollick Ethan Mollick is an Influencer
9mo

Google’s new Gemini AI model is out. The big deal is that Gemino Ultra appears to be the first model to beat GPT-4. The fascinating thing is that it does it by just a tiny bit. For right now, they are only integrating Gemini Pro into Bard. Gemini Pro is a GPT-3.5 class model, so it will not be up to the standards of GPT-4, yet.
3 Comments
Like Comment
To view or add a comment, sign in
Gayan Samuditha

PhD Candidate in Biomedical & Health Informatics | Advancing Healthcare through Data-driven Insights | Researcher, Former Senior Software Engineer & Visiting Lecturer
9mo
Report this post
Big Deal with Gemino Ultra
Ethan Mollick Ethan Mollick is an Influencer
9mo

Google’s new Gemini AI model is out. The big deal is that Gemino Ultra appears to be the first model to beat GPT-4. The fascinating thing is that it does it by just a tiny bit. For right now, they are only integrating Gemini Pro into Bard. Gemini Pro is a GPT-3.5 class model, so it will not be up to the standards of GPT-4, yet.
Like Comment
To view or add a comment, sign in
Eddie Lou

Active early stage and growth stage investor
2mo Edited
Report this post
Check out the most accurate AI Search engine! iAsk.Ai
Leonard Rodman, M.Sc. PMP® CSM® CSPO®

IT System Administrator | AI Implementation Analyst | Agile Project Manager | 38k followers & 11M views/14mo | ChatGPT, Midjourney, Runway and more!
2mo

I'm always looking for the top AI models to test out. Did you hear that www.iAsk.Ai Pro is the first model to achieve a "Superhuman" MMLU score? That means that it's currently not only beating models like GPT-4o and Claude 3.5 Sonnet, but it's also beating human experts! I'm sure this is going to be all over the news by later this year, but for now, you can get ahead of the curve and try this awesome product today at https://iask.ai/mmlu I'm really blown away by testing its performance even briefly so far! Don't forget to check it out at https://iask.ai/mmlu
1 Comment
Like Comment
To view or add a comment, sign in
Jun Heider

Tech Leadership and Innovation Strategist: Driving for Continuous Optimization and Visionary Solutions
3mo
Report this post
Decent and very concise summary on how to be less of an energy consumer while deriving benefits from AI language models. Bonus on running privately and on the edge as well. "Using Lightweight AI with Small Language Models" by Morten Rand-Hendriksen! Check it out: https://lnkd.in/gFaJM6mc #artificialintelligence, #largelanguagemodels, #llm, #slm

Certificate of Completion

linkedin.com

2 Comments
Like Comment
To view or add a comment, sign in

Alex Banks

98,167 followers

View Profile Follow

More from this author

Explore topics

By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.

翻译：