Local Phi-4 (14B) Test using Ollama - Summarization, Structured Text Extraction, Data Labelling

Phi-4 is Microsoft Research's latest iteration of the Phi family of models. It has 14B parameters, which is not small anymore. Can it beat Qwen 2.5 for your Ollama local setup? Let's test it!

Blog post: https://techcommunity.microsoft.com/b...
Technical Report: https://arxiv.org/abs/2412.08905
Ollama model: https://ollama.com/vanilj/Phi-4
Peter Lee tweet: https://x.com/peteratmsr/status/18673...

AI Bootcamp: https://www.mlexpert.io/
LinkedIn:   / venelin-valkov  
Follow me on X:   / venelin_valkov  
Discord:   / discord  
Subscribe: http://bit.ly/venelin-subscribe
GitHub repository: https://github.com/curiousily/AI-Boot...

👍 Don't Forget to Like, Comment, and Subscribe for More Tutorials!

00:00 - Welcome
00:10 - Phi-4 blog post
01:39 - Peter Lee tweet
02:24 - Technical report
03:16 - Ollama model
04:19 - Setup with Ollama
05:05 - Coding
07:09 - Data labeling
10:28 - Text summarization
10:53 - LinkedIn post
11:53 - Structured data extraction
13:14 - Rag/Question-answering
15:19 - Table data extraction
17:00 - Conclusion

Join this channel to get access to the perks and support my work:
   / @venelin_valkov  

#phi #ollama #chatbot #chatgpt #artificialintelligence #python