Notebooks
W
Weaviate Rag Gpt Oss 20b

Rag Gpt Oss 20b

vector-searchvector-databaseretrieval-augmented-generationopenaillm-frameworksweaviate-featuresfunction-callingweaviate-recipesmodel-providersPythongenerative-ai
alph-notebooks/weaviate-recipes / rag_gpt-oss-20b.ipynb

Open In Colab

Local RAG with OpenAI gpt-oss:20b via Ollama

In this demo notebook we will show how to use OpenAI gpt-oss:20b model via Ollama for Retrieval Augmented Generation (RAG).

Requirements

Ollama

Installation instructions can be found here.

  • Run the local generative model: ollama run gpt-oss:20b
  • Run the local embedding model: ollama pull snowflake-arctic-embed2

Local Weaviate instance

  • Steps for running Weaviate locally can be found here.
  • Don't forget to enable the text2vec-ollama and generative-ollama modules.

Install the Weaviate Python Client

[ ]

Configure Weaviate

[ ]

Setup Products Collection

[ ]

Load the Dataset

[ ]

Retrieval Augmented Generation w/ gpt-oss:20b

[ ]
[ ]