Perform k-NN similarity search and clustering on dense vectors with FAISS.
npx skills add https://github.com/zangjeicy/Hermes --skill faiss-zangjeicy
Please help me install this Agent Skill. Skill: faiss Source: https://github.com/zangjeicy/Hermes/tree/main/optional-skills/mlops/faiss Command: npx skills add https://github.com/zangjeicy/Hermes --skill faiss-zangjeicy
This skill addresses the challenge of performing fast, accurate similarity searches across massive collections of high-dimensional vectors, which is essential for modern AI applications like RAG and recommendation systems.
Use the faiss skill to initialize a new IndexFlatL2 index with 128 dimensions and add the provided vector dataset for similarity searching.
High-intent search queries and answers about installing and using this skill.
Similarity search on large-scale dense vectors is performed using Facebook's FAISS library, which supports k-NN search and approximate nearest neighbor indexing. It enables efficient retrieval across millions or billions of vectors using index types like HNSW and IVF.
Vector retrieval for RAG applications is accelerated using FAISS with optional GPU support. By leveraging hardware-accelerated similarity search, you can achieve 10-100x performance gains over CPU-only implementations while maintaining high throughput and low latency.
Yes, FAISS works seamlessly with numpy arrays for approximate nearest neighbor indexing. The skill depends on numpy alongside faiss-cpu and faiss-gpu to process dense vectors, allowing you to index and search datasets efficiently.
GPU acceleration provides significant performance improvements for dense vector clustering and similarity search. Using faiss-gpu delivers 10-100x speedups compared to CPU-only implementations, satisfying requirements for high-throughput vector operations on massive datasets.
HNSW and IVF are both supported index types for approximate nearest neighbor search in FAISS. Choosing between them depends on your specific latency and accuracy requirements when performing similarity search and clustering on large collections of dense vectors.
To initialize an IndexFlatL2 index for k-NN search, specify the vector dimensions, create the index, and add your dataset. This skill enables you to quickly set up a 128-dimensional index and perform exact similarity searches on dense vectors.
zechenzhangAGIPerform billion-scale k-NN similarity search with FAISS index types and GPU acceleration.
akrindevGenerate text embeddings and calculate semantic similarity via the Gemini Embedding API.
blueskies1818Perform k-NN vector similarity search and clustering with FAISS.
cxnaivePerform high-speed vector similarity search on large-scale datasets using FAISS.
