This repository contains some exercises done at the course Enterprise RAG and Multi-Agent Applications, offered by Hamza Farooq at Maven. For more information, please visit the course page.
- Exercise 1a - Semantic cache using Embeddings and FAISS vector DB
- Exercise 1b - RAG application using BigQuery and Redis
- FAISS - A library for efficient similarity search and clustering of dense vectors.
- Massive Text Embedding Benchmark (MTEB) Leaderboard - A leaderboard for text embedding models.
- Redis VL - Redis module for vector similarity search.
- RAG example application - Jupiter notebook with semantic caching using Couchbase and JINA embedding.
- Exercise 2a - Run locally Mistral-7B-Instruct
- Exercise 2b - Run previously application with GROK
- Exercise 2c - Fine-tuning with unsloth an Urdu chatbot
- Multi-task Language Understanding - A collection of papers, code and evaluation results for multi-task language understanding.
- LLM Qunatization and Inferencing Tool - A repository for quantizing large language models and inferencing quantized models from Hugging Face.
- Decoding Perplexity - A blog post on decoding perplexity and its significance in large language models.
- LM Studio - An application for downloading and running local LLMs.
- Are you GPU poor? - Calculate GPU memory requirement and token/s for any LLM.
- Language Model Evaluation Harness - A framework for few-shot evaluation of language models.
- Run Pod - A Globally distributed market GPU cloud.
- Vast.ai - Low-cost cloud GPU rental.