Llama 3.1 Review – A first look at Llama 405B and other updated Llama 3.1 models

preview_player
Показать описание
Llama 3.1 Review – A first look at Llama 405B and other updated Llama 3.1 models 👀

✅ In this quick video, we'll give you the 101 about Meta's brand new models and do a quick live demo.

🚀 Is Meta's Llama 3.1 the ultimate AI game-changer? Dive deep with Jordan Wilson from Everyday AI as we unpack Meta's latest breakthrough – Llama 3.1! Just minutes ago, Meta dropped Llama 405B, and we're bringing you a first-hand look at what this colossal model can do. From jaw-dropping benchmarks to impressive live demos, this video covers it all!

Curious about how Llama 3.1 stacks up against GPT-4 Omni or Claude 3.5 SONNET? We've got you covered! Jordan will break down the MMLU scores, reasoning abilities, and even some quick logic tests to show you why this model is the talk of the AI world. Plus, discover why Meta is shifting its focus from the Metaverse to generative AI and what this means for the future of AI technology.

Join us for an unfiltered, unscripted review of Meta's latest AI marvel. Don't miss out on this exciting reveal – hit subscribe, drop a "pancakes" in the comments if you watched till the end, and let us know what else you want to see about Llama 3.1! 🦙✨

#Llama31 #MetaAI #AIReview #GenerativeAI #BenchmarkTests #EverydayAI #JordanWilson #AIGamechanger #AiUpdates #LatestAiModels #MetaLlama31Review #MetaLlama31 #llama3 #llama31 #405B #llama405b #opensource #llm #llms #largelanguagemodels

CHAPTERS:
00:00 - Intro to Lllama 405B and Llama 3.1
1:06 - Meta LLM 3.1 Overview
2:30 - Performance Benchmarks
6:42 - Live Llama 405B Demo Analysis
9:31 - MetaLlama 3.1 Features
10:38 - MetaLlama 3.1 Logic Capabilities
12:59 - MetaLlama 3.1 Strategy Insights
14:13 - Switching to MetaLlama 3.1 70B
15:38 - Final Thoughts on Lllama and Open Source
16:50 - OUTRO
Рекомендации по теме
Комментарии
Автор

And just realized that it got the 3-digit code WRONG, not correct. Whoops.

EverydayAI_
Автор

gpt 3.5 was there to show that now its possible to run a better model than it locally in your computer. at least, i highly appreciate it was there for this exact motive

RafaelFleury_
Автор

MMLU is notoriously unreliable. Benchmarks aren't reliable, and Sonnet 3.5 is much better than GPT 4 omni at writing and code.

adamholter