Preparing the page…

Taking longer than expected? Reload

CATEGORY INTELLIGENCE / WORLDWIDE

Vector databases

The retrieval infrastructure behind AI products.

Red ocean · Measured classification

Vector databases: Search falling · established alternatives

Search attention for “vector database” is falling. 14 reviewed open-source projects serve this task. Established open-source alternatives shape this category. Compare their strengths and find a concrete reason for users to choose your product.

Report dated 2026-10-15 · Method 2.0.0 · moderate evidence confidence

Measured search term: vector database · Last 8 complete weeks vs previous 8

2832 active projects match the published GitHub search scope.
The keyword “vector database” has 104 complete weeks; 0% are reported as zero.

Open-source competition: ≥77 / 100 · Established alternatives
Open-source alternatives: 14
Direction basis: Recent search windows
Matching active GitHub projects: 2,832
Search-interest growth: -39% · Last 8 complete weeks vs previous 8
Search term and region: vector database · Worldwide
Complete weekly observations: 104

Why this classification

Median weekly search interest fell 39% across two consecutive eight-week windows.
2,832 active repositories match this search scope. Review their users and workflows to identify direct alternatives.
Four-week search change: -46%; thirteen-week change: -15%. These windows check the direction of the eight-week comparison.
Project roles: 14 direct alternatives, 0 adjacent projects, 1 resources, 85 awaiting review.
Competition pressure: 77–100/100; breadth 27, established alternatives 30, leading project strength 20.

Source evidence

GitHub repository search · topic:vector-database fork:false archived:false stars:>=1 pushed:>=2025-10-15

Collected 2026-10-15T04:47:35.086Z

Google Trends search interest · Collected 2026-10-15T04:47:35.086Z

Leading repositories

Repository	Stars	Description
Mintplex-Labs/anything-llm	67,103	Stop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience
meilisearch/meilisearch	59,571	A lightning-fast search engine API bringing AI-powered hybrid search to your sites and applications.
pathwaycom/serviette	58,789	Ready-to-run cloud templates for RAG, AI pipelines, and enterprise search with live data. 🐳Docker-friendly.⚡Always in sync with Sharepoint, Google Drive, S3, Kafka, PostgreSQL, real-time data APIs, and more.
run-llama/llama_index	52,501	LlamaIndex is the document processing platform for AI
milvus-io/milvus	46,383	Milvus is a high-performance, cloud-native vector database built for scalable vector ANN search
VectifyAI/PageIndex	39,129	📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG
qdrant/qdrant	35,079	Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
topoteretes/cognee	32,312	Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory with small models for free
NirDiamant/RAG_Techniques	29,874	This repository showcases various advanced techniques for Retrieval-Augmented Generation (RAG) systems. Each technique has a detailed notebook tutorial.
weaviate/weaviate	16,884	Weaviate is an open-source vector database that stores both objects and vectors, allowing for the combination of vector search with structured filtering with the fault tolerance and scalability of a cloud-native database.

Limits of this result

Google Trends measures relative search attention. Assess customers, revenue and willingness to pay through direct research.
Supply covers active repositories matching the displayed GitHub queries. Broader competition research can include other repositories and commercial products.
Repository density and search attention provide complementary research signals. Assess competition and demand through user and product evidence.
Competition pressure is an operational index of observed open-source alternatives. Stars and forks indicate developer attention and reuse; user adoption and commercial products deserve separate research.
The displayed projects form a stars-ranked sample. Competition pressure is a lower bound; wider coverage can strengthen the assessment.
Some project roles await closer review. The pressure range includes their possible contribution.

Search interest measures attention. Customer demand needs user research; classification thresholds are published heuristics subject to empirical calibration.

Community requests

Windows support / Binaries for Windows

Discussion open

weaviate/weaviate · Posted: 2023-07-31 · Updated: 2025-01-30 · Collected: 2026-10-15

[Enhancement]: Allow cluster-local Flush and FlushAll on both primary and secondary without CDC replication

Discussion open

Is there an existing issue for this? - [x] I have searched the existing issues What would you like to be added? Decouple cluster-local, non-data-mutating control APIs from primary/secondary replication admission and CDC forwarding, starting with Flush and

milvus-io/milvus · Posted: 2026-10-15 · Updated: 2026-10-15 · Collected: 2026-10-15

Support for batch patch (partial update)

Discussion open

I need (quite often) to do updates on some of the columns/properties of my class... But there is no batch update functionality... batch update one-by-one take forever... For example update of 400k records takes about 20h... which is crazy long... (I can

weaviate/weaviate · Posted: 2022-08-22 · Updated: 2025-08-28 · Collected: 2026-10-15

Feature Request: Stemming on Inverted Index text props

Discussion open

When a text such as "The race cars are queued up at the pit exits" is indexed, it cannot be found using exact filters or BM25/Hybrid when the query is "race car queue at pit exit" because of car!=cars, queue!=queued, exit!=exits, etc. To fix this, the word

weaviate/weaviate · Posted: 2022-12-14 · Updated: 2023-09-01 · Collected: 2026-10-15

Support global vector index based on clustering compaction

Discussion open

Motivation Vector clustering compaction already reorganizes data by similarity. This work extends centroid pruning by publishing the cluster layout and its index as one complete PartitionStatsSnapshot, so queries can translate cluster candidates into physica

milvus-io/milvus · Posted: 2026-08-04 · Updated: 2026-10-14 · Collected: 2026-10-15

Feature request: Cross-tenant queries

Discussion open

Feature description Ability to search vectors across multiple tenants in a single query. Use case Imagine a scenario where different tenants grant partial access to data to a user. For this user to search all the documents they were granted access to mu

weaviate/weaviate · Posted: 2023-09-04 · Updated: 2026-04-22 · Collected: 2026-10-15

FR: Support SPLADE / Sparse vectors

Discussion open

Describe your feature request There have been quite a few comments lately about supporting SPLADE. Here's an explicit feature request. If you want/need SPLADE, please upvote this. Code of Conduct - [X] I have read and agree to the Weaviate's [Contributor

weaviate/weaviate · Posted: 2024-01-16 · Updated: 2025-01-14 · Collected: 2026-10-15

[Enhancement]: Eliminate post-import SortCompaction by reworking the import execution pipeline (Import V3 reshard)

Discussion open

Is there an existing issue for this? - [x] I have searched the existing issues What would you like to be added? A reworked import execution path ("Import V3 reshard") where source files are parsed and sorted once at first read, and final segments are prod

milvus-io/milvus · Posted: 2026-09-09 · Updated: 2026-10-14 · Collected: 2026-10-15

[Proposal] Add [ImageRef] image reference type

Discussion open

Hi team, I would like to add a new type to Weaviate: ImageRef. The problem Currently, when using image vector search, we need to store a base64 representation of each image in a blob property. However, in most databases, you rarely would want to query y

weaviate/weaviate · Posted: 2023-03-17 · Updated: 2025-10-28 · Collected: 2026-10-15

[Feature]: Modify the collection schema once collection is created and is not empty

Discussion open

Is there an existing issue for this? - [X] I have searched the existing issues Is your feature request related to a problem? Please describe. Many a times, when we start with an ANN collection definition we don't know the exhaustive list of fields which s

milvus-io/milvus · Posted: 2022-11-08 · Updated: 2026-09-20 · Collected: 2026-10-15

Allow log timestamps to follow TZ / local timezone

Discussion open

Feature request Please allow Qdrant log timestamps (stdout / docker logs) to follow the container timezone, for example via the TZ environment variable or a dedicated logger config option. Current behavior - Official Docker images set TZ=Etc/UTC and install

qdrant/qdrant · Posted: 2026-10-13 · Updated: 2026-10-14 · Collected: 2026-10-15

FR: Allow arbitrary ID instead of forcing a UUID

Discussion open

Hi, Weaviate requires the predefined uuid field 'id' to be the key for all objects. It is used for instance to delete or update objects. But 100% applications have their own primary key (SQL, WordPress, ...), and most often it is not a uuid4 formatted t

weaviate/weaviate · Posted: 2021-09-19 · Updated: 2025-08-28 · Collected: 2026-10-15

Use and share the evidence

Permanent report · Markdown · JSON · PNG card

Use the open-source CLI and MCP server on GitHub