Embeddings & Vector Search

Company-reported MTEB Code rose from 68.76 to 78.68; context is 8K (4 times its predecessor), and MRL lets the 768 dimensions shrink to 128, cutting storage by up to 6 times. Quantized, it runs on a Pixel 11 Pro in about

A case of the competitive focus of embedding models moving from scores to operating costs. The design in which tiers share an embedding space aims to eliminate reindexing costs — the biggest barrier to switching models i

The point at which data warehouse queries became a basic agent tool without going through analysts, showing the BI tool layer being absorbed into agents.

A strand that claims multi-level ranking supervision makes better use of the supervision signal than contrastive learning, long the default for embedding training, and repositions the reranker not as the back end of the

A strand that, instead of the ColPali lineage of reusing generative models as retrievers, returns to a single tower designed as an encoder from the start — a claim that "encoder-only" is valid again for multimodal retrie

An attempt to absorb chunking, a RAG preprocessing step, into the embedding model itself

An integrated search engine using quantization to take direct aim at the performance and cost case for dedicated vector DBs

The commercial embedding lineage officially moved past the text-only generation into a natively multimodal one

The counterargument of the dedicated vector DB camp. A direct reply to integrated options (Postgres, Elasticsearch)

Applies MoE to commercial embeddings and standardizes embedding compatibility across models of different sizes. Comparison figures are the vendor's own measurements

Showed that, right after the view that embeddings have limits spread, the commercial importance of the reranking layer actually grew

An attempt to absorb embedding generation and ANN tuning into the database and eliminate separate pipelines

A cloud provider makes unified multimodal embeddings a standard product

A rebuttal showing that the limits of single-vector retrieval are fundamental, dimensional limits that more data or bigger models cannot fix. A theoretical basis for a return to multi-vector, sparse and hybrid retrieval

Rerankers change from fixed functions into components steered by instructions. Comparison figures are the vendor's own measurements

The point at which quantized indexes moved from an option to the default

Absorbs chunking, a fundamental flaw of RAG, into the model itself

Commercialized the experimental version from March, entering the embedding API race in earnest

Points toward a single model handling both multimodality and late interaction

The axis of vector DB competition moved from adding features to cutting the cost of billion-scale vectors

The point where open-weight embeddings overtook commercial API models on the leaderboard

The point the open-weight camp reached by releasing embeddings and rerankers as a single lineage

An attempt to redefine the category from vector index to AI data lakehouse

A radical approach to the cost of migrating embeddings, which also raised security implications for leaked vector DBs

Commercial embedding competition moved beyond text search to long documents and mixed-modality search

A commercial multimodal embedding that handles text, images and document screenshots in one model. The commercialization of the direction ColPali opened

Represents the shift of embedding models from small dedicated encoders to derivatives of frontier LLMs

A general-purpose database vendor directly bought a company making embedding and reranking models. A decisive phase in the contest between dedicated vector DBs and extensions of existing databases

The de facto default vision encoder for VLMs. Multimodal embeddings and VLM backbones converged into the same lineage

The community's response to overfitting on the MTEB leaderboard. Moved the baseline for the embedding rankings race

The phase in which competition among commercial embedding APIs explicitly took aim at the OpenAI baseline

A sign that data warehouse vendors were starting to build embedding models themselves

The point at which rerankers were commercialized as a separate foundation-model product line. The two-stage structure settled in as the default architecture for RAG

The first competitive embedding model to release even its training data. Added reproducibility as a new dimension of the embedding race

MRL-style truncation became a built-in feature of a commercial API

Secured capital for the approach of a Rust-based, high-performance, single-purpose engine

Shifted the cost structure of vector DBs from always-on to usage-based

Sparked the race in long-document embeddings. The first open model family to reduce the burden of chunking

An open-weight strategy of releasing benchmark, data and models as one bundle changed the landscape of the MTEB leaderboard

HNSW became usable in an existing RDB without a dedicated vector DB, directly affecting the shape of the vector DB market

A sign that small open-weight models were starting to overtake commercial APIs

A representative case of capital flowing into vector DBs at the time of the 2023 RAG boom

The event that created the category of embedded vector stores for LLM app developers

The de facto successor to CLIP in multimodal embeddings. Removing the need for very large batches greatly lowered the barrier to training

Turned embeddings into a commodity and removed the cost barrier to the RAG boom

A rebuttal to the challenge BEIR had raised. The starting point of the open-weight embedding race

The common yardstick for the embedding model race. Its key conclusion: no embedding method dominates every task

A technique that made embedding dimensions adjustable after the fact

The point at which existing search engines began absorbing vector search. The starting point of the consolidation debate over whether a dedicated vector DB is needed

The strand that made lexical retrieval learnable by neural networks. Broke down the dense-versus-sparse dichotomy

An event showing that the vector DB could stand as foundation-governed open-source infrastructure rather than a vendor product

The start of the path that brought vector search into existing Postgres rather than a separate DB

A decisive rebuttal of the narrative that dense retrieval had beaten lexical retrieval. The practical grounds for hybrid search becoming the default

The first commercialization of the vector DB as a standalone product category. Triggered the market formation of Weaviate, Qdrant, Chroma and Milvus

Redesigned the quantization loss function around inner-product estimation error, pushing the quality-speed frontier of MIPS forward by a generation

Flipped the default from sparse retrieval (BM25) to dense retrieval

The first study to tackle the cost of embedding migration head-on. Framed as a problem the structural debt of having to re-embed the entire corpus every time the model is replaced

An SSD-based ANN that broke HNSW's constraint of having to hold everything in memory

Turned sentence embeddings into a searchable asset

The prototype of the reranker lineage. Standardized the two-stage structure that recovers, at a later stage, the precision dense embeddings miss

The library that brought large-scale vector search into the realm of practical use

The default indexing algorithm of almost every vector DB today

The starting point of every dense retrieval and embedding stack that followed

The origin point of the IVF-PQ family. The compressed indexes of FAISS, ScaNN and most vector DBs descend directly from this paper