pgvector HNSW on a small VPS: how much RAM?
Size a pgvector HNSW index before you build it: the byte arithmetic, maintenance_work_mem, ef_search, and what to do when it stops fitting in RAM.
#rag
Filtering by topic #rag · clear
Run EmbeddingGemma 2 on a CPU-only VPS
Ollama 0.40 cannot run EmbeddingGemma 2 on a Linux CPU yet. Here is the sentence-transformers setup that does, with task prefixes and 256d vectors in pgvector.
Build your own AI agent on a VPS
An AI agent is a loop around a language model that can use tools. Learn the concepts behind building one on your own VPS: the loop, tools, MCP, memory.
Running a vector database on your VPS
Your app and your index sit on the same VPS, so latency is not the problem. Compare pgvector, Qdrant, Chroma and brute force, and size the RAM correctly.
Build a RAG pipeline on your own VPS
Chunk, embed, store and retrieve on one VPS. A pgvector schema, HNSW sizing, a local embedding model, and the SQL that proves retrieval works.