# TopK

> The search engine beyond vector search, keyword search, and rank fusion.

Hybrid search, multi-vector retrieval, custom ranking, and managed inference in one API. Built on object storage for 10x lower cost and massive scale.

## Overview

- [Home](https://www.topk.io): The retrieval engine for search, RAG, and agents. Hybrid search, multi-vector retrieval, custom ranking, and managed inference in one API.
- [Documentation](https://docs.topk.io): Product documentation, Python and JavaScript/TypeScript SDK references, CLI docs, and MCP server setup.
- [Quickstart](https://docs.topk.io/quickstart): Get started with TopK in minutes and run your first hybrid search.
- [Docs llms.txt](https://docs.topk.io/llms.txt): Machine-readable index of TopK documentation pages.

- [Console](https://console.topk.io): Manage your account and billing, get API keys, and monitor usage.
- [Pricing](https://www.topk.io/pricing): Usage-based pricing for ingesting data and querying TopK.
- [Benchmarks](https://www.topk.io/benchmarks): End-to-end benchmark results: query latency (p50/p95/p99), throughput, ingest time, and filtered-search performance from 1M to 1B vectors, plus File Search answer accuracy across finance, legal, medical, and industrial domains.
- [FAQ](https://www.topk.io/faq): Frequently asked questions about TopK — what it is, how it compares to alternatives, and how to get started.
- [Blog](https://www.topk.io/blog): Engineering updates, product announcements, benchmarks, and retrieval research from the TopK team.
- [Answers](https://www.topk.io/learn): Quick, definite answers to specific questions about advanced retrieval: multi-vector search, MaxSim, hybrid ranking, and RAG for agents.
- [Team](https://www.topk.io/team): Information about the TopK team, hiring, and company culture.
- [Contact](https://www.topk.io/contact): Contact form and support email for sales, support, and general inquiries.

## Blog Authors

- [Jergus Lejko](https://www.topk.io/blog/author/jergus-lejko): Blog posts by Jergus Lejko.
- [Marek Galovic](https://www.topk.io/blog/author/marek-galovic): Blog posts by Marek Galovic.
- [Martin Spisak](https://www.topk.io/blog/author/martin-spisak): Blog posts by Martin Spisak.
- [TopK Team](https://www.topk.io/blog/author/topk-team): Blog posts by TopK Team.

## Blog Posts

- [RAG Is Broken for Agents. Here's How We Fixed It.](https://www.topk.io/blog/20260628-rag-is-broken-for-agents): Context is a search problem. Without the right context, even the best models fail. This post describes why dense embedding based RAG is broken for agents and how multi-vector (late interaction) retrieval fixes it.
- [TopK SQL: A Search Query Language](https://www.topk.io/blog/20260614-topk-sql): TopK now implements the Postgres wire protocol, so any Postgres client can run semantic search, hybrid search, and filtered retrieval as ordinary SQL.
- [High-Quality Search, Out of the Box](https://www.topk.io/blog/20260611-semantic-index-multi-vector-retrieval): TopK's semantic_index annotation brings state-of-the-art multi-vector retrieval to production — no embedding pipeline, no separate vector store, no reranking service.
- [SMVE: Multi-Vector Retrieval That Just Works](https://www.topk.io/blog/20260311-smve-multi-vector-retrieval): TopK introduces SMVE, a fast and scalable approach to late-interaction retrieval based on sparse random projections.
- [TopK Bench: Benchmarking Real-World Vector Search](https://www.topk.io/blog/20251201-topk-bench): This post evaluates how managed vector databases handle large-scale, production-like workloads—including millions of vectors, concurrent querying, real-world filtering, and continuous ingestion—using reproducible benchmarks across several leading providers.
- [[Podcast] Building a Search Database from First Principles](https://www.topk.io/blog/20250907-podcast-building-search-database-first-principles): Listen to our founder Marek Galovic discuss the challenges and solutions in building a search database from the ground up, covering everything from vector search to hybrid retrieval and the architectural decisions that shaped TopK.
- [Beyond RRF: How TopK Improves Hybrid Search by up to 7.8%](https://www.topk.io/blog/20250724-beyond-rff-how-topk-improves-hybrid-search-quality): A case study on how TopK hybrid retrieval outperforms traditional RRF-based methods, improving nDCG@10 by an average of 4.5% (and up to 7.8% on some datasets) by leveraging score-aware ranking, dynamic weighting, and efficient result merging.
- [Scaling Without Complexity: Billion-Scale Hybrid Search with TopK](https://www.topk.io/blog/20250722-billion-scale-hybrid-search): TopK enables billion-scale hybrid search with <100ms latency, fast indexing, and high-quality results.
- [Binary Vector Search at 350GB/s using ARM NEON](https://www.topk.io/blog/binary-vector-search-arm-neon): Optimizing binary vector search using ARM NEON instructions to achieve 350GB/s throughput.
- [We Raised $5.5 Million to build an AI-Native Search Engine for Enterprises](https://www.topk.io/blog/seed-round): TopK has raised $5.5 million in seed funding to accelerate the development of its unified, AI-native search platform, enabling organizations to seamlessly combine vector, keyword, and custom ranking in a single system. The round was led by top-tier investors, fueling our mission to redefine search for the AI era.
- [TopK at Data Council AI Launchpad 2025](https://www.topk.io/blog/topk-at-data-council-2025): At Data Council 2025, we introduced TopK, a unified cloud-native query engine that combines vector search, keyword matching, and flexible scoring in a single system. We demonstrated how TopK overcomes the limitations of traditional vector databases by supporting hybrid queries with custom ranking logic while maintaining high performance.
- [Why Vector DBs Are the Wrong Abstraction – And What We Built Instead](https://www.topk.io/blog/vector-dbs-are-the-wrong-abstraction-how-we-built-a-new-search-database-from-scratch): We’ve spent the last three years building the most popular vector database on the market. In that time we realized that a database built around vectors as a primary key is simply the wrong abstraction, creating an unnecessary obstacle for users in production.

## Answers

- [Can You Convert Vectors Between Models?](https://www.topk.io/learn/converting-embeddings): No. Different models produce incompatible spaces, and learned mappings lose the distinctions you upgraded for. Plan to re-embed, and make it cheap.
- [Can You Run Vector Search in SQL?](https://www.topk.io/learn/sql-vector-search): Yes, two ways: extensions like pgvector add distance operators to Postgres, and TopK speaks the Postgres protocol so hybrid search is ordinary SQL.
- [Collections or Filters for Multi-Tenancy?](https://www.topk.io/learn/multi-tenant-vector-search): Default to one shared collection with a mandatory tenant filter; split out only huge or contractually isolated tenants. Idle-collection cost decides.
- [Do You Still Need a Reranker?](https://www.topk.io/learn/reranking): Add a reranker when the right documents reach your top 100 but not your top 5. With a multi-vector first stage, measure first: it may be redundant.
- [Does Quantization Hurt Accuracy?](https://www.topk.io/learn/vector-quantization-accuracy): Yes, but recoverably: quantize for candidate selection, then rescore the shortlist at full precision. The loss lands in the tail, not the top results.
- [How Do You Audit an AI Answer?](https://www.topk.io/learn/auditable-ai-answers): An answer is auditable when every claim cites a document and page a human can open. That requires citations from retrieval, not citations written by the model.
- [How Do You Benchmark Vector Search?](https://www.topk.io/learn/benchmarking-vector-search): Benchmark your own workload: mixed read/write load, your filters, and latency percentiles at a fixed recall target. Read-only QPS flatters every engine.
- [How Do You Chunk Redundant Documents?](https://www.topk.io/learn/chunking-redundant-documents): Deduplicate before you embed: canonicalize versions, hash-dedupe chunks, and reach history through filters. Indexing every version multiplies cost.
- [How Do You Enforce RBAC in Vector Search?](https://www.topk.io/learn/rbac-vector-search): Store access metadata on every document and enforce entitlements as a query-time filter in the engine. Post-filtering leaks; per-role indexes explode.
- [How Do You Upgrade Embedding Models?](https://www.topk.io/learn/upgrading-embedding-models): Blue-green the index: backfill with the new model, dual-write, shadow-test, flip reads, keep the old index for rollback. Model vectors never mix.
- [How Should You Weight Hybrid Scores?](https://www.topk.io/learn/hybrid-score-weighting): There is no single right weight: short queries lean lexical, verbose ones lean dense. Condition weights on the query, or use RRF and skip them.
- [Is Multi-Vector Retrieval Too Expensive?](https://www.topk.io/learn/multi-vector-at-scale): It was: token-level embeddings cost 10–100× more storage. Two-stage designs and MaxSim approximations like SMVE now make it a viable first stage.
- [RRF vs Score Fusion vs True Hybrid: Which Should You Use?](https://www.topk.io/learn/rrf-vs-true-hybrid): RRF merges lists by position, score fusion tunes weighted scores, and true hybrid scores every signal in one query. Start with RRF; move up as you measure.
- [Top-k (Retrieval)](https://www.topk.io/learn/top-k): Top-k means returning the k highest-scoring results from a ranked search or model. It is a ranking cutoff, distinct from TopK, the search engine.
- [What Is MaxSim?](https://www.topk.io/learn/maxsim): MaxSim sums each query token's best match against a document's tokens. It is the scoring operator behind multi-vector (late-interaction) retrieval.
- [What Is Multi-Vector Retrieval?](https://www.topk.io/learn/multi-vector-retrieval): Multi-vector retrieval stores one embedding per token and scores documents with MaxSim, so precise queries match exact passages instead of a lossy document summary.
- [When Do You Need Hybrid Search?](https://www.topk.io/learn/hybrid-search): Use hybrid search when queries mix natural language with exact tokens like names and codes. Vectors blur exact terms; keywords miss paraphrases.
- [When to Use Multi-Vector Embedding Models](https://www.topk.io/learn/when-to-use-multi-vector-embeddings): Use multi-vector models when queries target specific details in long or complex documents; use single-vector when queries are broad or cost dominates.
- [Which Recall Number Actually Matters?](https://www.topk.io/learn/recall-vs-retrieval-quality): Index recall measures whether ANN approximates exhaustive search. Retrieval quality measures whether the right document surfaces. Evaluate quality first.
- [Why Do Upserts Slow HNSW Reads?](https://www.topk.io/learn/hnsw-upsert-performance): Reads and writes contend on one graph: inserts rewire lists mid-traversal and deletes leave tombstones. Sustained upserts tax every read.
- [Why Do Vector Databases Re-Shard?](https://www.topk.io/learn/shard-rebalancing): Stateful engines pin data to nodes, so growth forces risky migrations. Storage-compute-separated engines pin nothing; there is nothing to rebalance.
- [Why Does Filtering Break HNSW?](https://www.topk.io/learn/filtered-vector-search): The graph was built without knowing your filter. Post-filtering empties results; in-traversal filtering breaks connectivity. Native filtering fixes both.
- [Why Does RAG Fail for Agents?](https://www.topk.io/learn/rag-for-agents): Agents fire many precise, long-tail queries: the worst case for single-vector RAG, which summarizes away the details. Fix retrieval, not the model.
- [Why Does RAG Fail on Financial Tables?](https://www.topk.io/learn/rag-financial-tables): Single-vector embeddings compress a whole page into one summary, so the row you need averages away into noise. Token-level retrieval matches at the cell.
- [Why Does RAG Retrieve the Wrong Chunks?](https://www.topk.io/learn/rag-wrong-chunks): Wrong chunks come from four places: chunk boundaries, lossy embeddings, missed exact tokens, and filters that empty results. Your failed queries tell you which.
- [Why Object Storage for Vector Search?](https://www.topk.io/learn/object-storage-vector-search): At billions of vectors the dominant cost is keeping data hot. Object storage decouples storage from compute, so cost tracks query load, not data size.

## Optional

- [Security](https://www.topk.io/security): Responsible disclosure and bug bounty policy for TopK.
- [Trust Center](https://trust.topk.io): Security compliance and trust information for TopK.
- [Privacy Policy](https://www.topk.io/privacy-policy): Privacy policy
- [Terms of Service](https://www.topk.io/terms-of-service): Terms of service