In this video, we'll delve into applying Google's new Gemma 2 model to create a simple PDF retrieval-augmented generation (RAG) system using the free version of Google Colab. We'll run the Gemma 2 model through Ollama and utilize Nomic's Embed text tool to create embeddings, which we'll store in ChromaDB.
Gemma 2 is available in 9 billion (9B) and 27 billion (27B) parameter sizes, offering enhanced performance and more efficient inference compared to its predecessor, along with significant safety advancements. The 27B model delivers competitive performance comparable to models over twice its size.
#ai #llm #localllm #generativeai #opensourceai
Blog:[ Ссылка ]
LINKS:
Code :[ Ссылка ]
[ Ссылка ]
[ Ссылка ]
[ Ссылка ]
