Meta Llama 3.3 70B Instruct Hot Take and Review - Video

Meta's new Llama 3.3 model is out, in 70B instruct format - and it's REALLY good.

For those unfamiliar, Llama is the family of AI models that Meta makes which compete with OpenAI, Google, and Anthropic. Meta's models are different in that you use your own service provider like IBM WatsonX to serve it, rather than have an AI company serve it for you.

That in turn allows you to have more data privacy. If you have a beefy enough laptop or server, you can even run it locally on your own hardware and then you don't even need Internet access.

What makes Llama 3.3 amazing is how good it is at following instructions. I ran a test against Llama 3.1 405B (their big state of the art foundation model that's competitive with ChatGPT) and Llama 3.3 outperformed its predecessor very, very well. Watch in the video as I put the two against each other in IBM WatsonX.ai (my preferred environment for running big local models because IBM's privacy is second to none).

Congratulations to Armand Ruiz, Libby Ingrassia and the team at IBM for having Llama 3.3 available for use within an hour of Meta's announcement.

hashtag#AI hashtag#GenerativeAI hashtag#GenAI hashtag#ChatGPT hashtag#ArtificialIntelligence hashtag#LargeLanguageModels hashtag#MachineLearning hashtag#IntelligenceRevolution hashtag#IBMChampion hashtag#IBMPartner

Subscribe to my weekly #email newsletter:
http://www.christopherspenn.com/newsletter

Please subscribe to my YouTube channel for more #marketing and #analytics videos!
https://www.youtube.com/user/christopherspenn

Need help with your company's #data, #AI, and #analytics? Let me know:
https://www.trustinsights.ai/contact

Join my free private Slack group, Analytics for #Marketers:
https://www.trustinsights.ai/analyticsformarketers

Take my new generative AI course:
https://www.trustinsights.ai/aicourse

I use generative AI to summarize and caption my YouTube videos.