New Llama 3.1 is The Most Powerful Open AI Model Ever! (Beats GPT-4)

Показать описание

Meta has unveiled Llama 3.1, the world’s largest open-source AI model with 405 billion parameters, challenging top models like GPT-4. This revolutionary AI model, trained on 16,000 Nvidia H100 GPUs, sets new benchmarks in AI capabilities and efficiency. Meta’s open-source approach allows developers to customize and enhance Llama 3.1 for various applications, ensuring widespread and innovative use across the industry.

#ai #meta

Рекомендации по теме

Комментарии

Yes, geopolitics is important to consider.
I think Meta are right in thinking open source will speed up AI development and make it safer by democratising it properly.

vincent_hall

very insightful and thoughtful commentary on open source ai. thank you.

atothezeezee

I like that Meta named their AI - Lama. It's cute. I like Groq. I enjoy Groq's edgy personality. When do we get voice inputs? When can we talk to AI and have it create video games and movies for us?

SHAOLINCOWBOY-iphw

I like Llama but it's the only model that when you talk with it in other languages it switch to English on it's own, i need to test 3.1 if there is progress

chanpasadopolska

So you're telling me that it's very powerful I couldn't understand the word of that but it sounds pretty good

ringerdiggsycobbsy

Glad its open source. Not sure it matche Claude though would have to see some benchmarks dor sure.

kingkura

We'll need a de-biased version, like Dolphin Ollama.

vincent_hall

Guys, relay this to ai developers, they are 🙃😦🦵👳👰about to code LSD into ai engines . So they should set this to reach ASAP !

Rhyth-hw

What if benefit not only uuge roups but and individuals, by making focus on it

ehjoepr

I hate to be that guy, but the video thumbnail is an elephant

amadzarak

If you’d stop slobbing Zuck’s knob for a minute and read the license terms, you’d find you have to get a license if your application is used 700M or more per day. It ain’t free.

SeattleShelby

Soo ... no mindflayer cyborg robo elephant-squids? Damn cause thats why i came here. Not even a Winamp joke here for crying out loud.

reptoid

You can trust all the open source releases cause we will be destroying them with our gaint corp AI
is what you meant to say.

rRobertSmith

Azure charge 0.000155 for Lama 3.1 and less for GPT-4o 0.00015❓❓❓

PromptEngineer_ChromeExtension

If it's 'open', where's the link to it? Also, you can only login using a fb or instagram account; hardly the definition of 'open'.

lucasmichaud-acapulco

New Llama 3.1 is The Most Powerful Open AI Model Ever! (Beats GPT-4)

Zuck's new Llama is a beast

Llama 3.1 Is A Huge Leap Forward for AI

Llama 3.1 is ACTUALLY really good! (and open source)

LLaMA 405b Fully Tested - Open-Source WINS!

Llama 405b BEAST already exploited | Here’s how

Metas LLAMA 405B Just STUNNED OpenAI! (Open Source GPT-4o)

LLaMA 3 Tested!! Yes, It’s REALLY That GREAT

Llama 405b: Full 92 page Analysis, and Uncontaminated SIMPLE Benchmark Results

Llama-3.1 Storm 8B - Improved SLM with Self-Curation + Model Merging

LLaMA 405b is here! Open-source is now FRONTIER!

LLaMA 3 UNCENSORED 🥸 It Answers ANY Question

EASILY Train Llama 3.1 and Upload to Ollama.com

This Llama 3 is powerful and uncensored, let’s run it

Bringing Llama 3 to Life | Joe Spisak, Delia David, Kaushik Veeraraghavan & Ye (Charlotte) Qi

How Did Llama-3 Beat Models x200 Its Size?

Metas LLAMA 3 Just STUNNED Everyone! (Open Source GPT-4)

Build Anything with Llama 3.1 Agents, Here’s How

Meta Announces Llama 3 at Weights & Biases’ conference

Mark Zuckerberg on Llama 3.1, Open Source, AI Agents, Safety, and more

Meta Llama 3 Schlägt sie Alle! – Das Beste OpenSource KI-Modell

Build Anything with Llama 3 Agents, Here’s How

Mark Zuckerberg - Llama 3, $10B Models, Caesar Augustus, & 1 GW Datacenters

LLAMA-3.1 🦙: EASIET WAY To FINE-TUNE ON YOUR DATA 🙌

'okay, but I want Llama 3 for my specific use case' - Here's how