Awesome LLM Experiments

Hands-on Python examples for LLM apps: Retrieval-Augmented Generation (RAG), embeddings, vector search, Chroma, LangChain, multilingual & multimodal (text, image) retrieval, and RAGAS evaluation.

Setup

Prerequisites

• Python 3.12+ (see pyproject.toml)
• uv for install and run

Install

uv sync

Run a notebook from the repo root or the experiment folder (see each experiment’s notebook for EXPERIMENT_DIR resolution).

Experiments

Experiment	Description	Notebook
multilingual-rag-retrieval	Bilingual UDHR (English + Hindi): E5 embeddings, Chroma vector DB, optional lang filters and cross-lingual search	notebook.ipynb
multimodal-text-image-vl-embeddings	Colab-only: cross-modal text+image VL embeddings with Qwen3-VL (Sentence Transformers). Open from GitHub in Colab, enable GPU runtime — not via uv sync	notebook.ipynb
ragas-rag-evaluation	English UDHR RAG evaluated with RAGAS: retrieval metrics (context precision/recall) + generation metrics (faithfulness/answer relevancy); MiniLM + Chroma + OpenAI	notebook.ipynb

Support

If you find this project helpful:

• ⭐ Star the repository
• 🐛 Report issues
• 🔀 Submit pull requests
• 💝 Sponsor on GitHub

License

MIT © Y. Siva Sai Krishna — see LICENSE for details.

---

Author's GitHub • Author's LinkedIn • Author's site • Report Issues