Tarık Şenkal Tags
Infrastructure
2 posts
Infrastructure7 minSpeculative Decoding: Getting Faster Tokens From the Same Model
Draft models, EAGLE, MTP, n-gram and suffix decoding. Why acceptance rate is the only metric that matters, how to configure vLLM, and when speculation does not help.
Infrastructure5 minVector Databases Guide: Similarity Search, HNSW, and FAISS / pgvector / Qdrant Compared
An intuitive guide to vector databases: similarity search, ANN and the HNSW algorithm, a FAISS vs pgvector vs Qdrant comparison, and when to choose each one.