“Weaviate is amazing! It lets us manage the data at scale when we need and how we need it.”
Vector DatabaseMaker says so · source ↗
Weaviate
Open-source vector database combining vector, hybrid, and keyword search for AI and RAG applications.
16 makers' products we track use Weaviate. For vector database it's 3rd most chosen of 7 options: 16 of the 107 makers' products we know the vector database of (15%). 32 open-source projects also declare it in their code. What makers like most: Open source and free to run, with good docs and an active, friendly community. Their most common complaint: Search can be slow, with metadata filtering hurting vector query performance and hosted retrieval latency complaints.
Profile checked
Works with AI agents:llms.txtHow makers use Weaviate
Makers' products we track use Weaviate for vector database (16), most often alongside SendGrid (50%), Vercel (44%).
What Weaviate is used for
Weaviate vs the alternatives makers chose
For vector database, 16 of the 107 makers' products we know the choice of picked Weaviate (15%), behind Pinecone (50%).
| Option | Makers' products | Free for | At 10,000 users | Open source | |
|---|---|---|---|---|---|
| 53 50% | ~3,300 users | $113/mo · Standard | No | Compare → | |
| 18 17% | ~1,000 users | $103/mo · Standard (3 nodes, 0.5 vCPU / 4 GiB each) | Yes | Compare → | |
| 16 15% | ~1,000 users | $253/mo · Flex | Yes | ||
| 12 11% | ~760 users | $51/mo · Starter | Yes | ||
| 8 7% | None | $21/mo · Launch | No | ||
| 3 3% | — | — | Yes |
For vector database: how many of the 107 makers' products we know the vector database of chose each option, and what each costs at 10,000 users.
Costs are the cheapest plan that fits, from each vendor's pricing page, at the usage our stack planner assumes for a product with that many users. Free for: the users its cheapest plan a business can use stays free for.
Often paired with
Share of the 16 makers' products on Weaviate that show this choice.When to use Weaviate, and when not
Pick Weaviate if
- You need an open-source store with heavy filtering or hybrid search, self-hosted or managed. Vector Database →
Something else fits better if
- You already use Postgres and have up to a few million vectors: pgvector.
- You're prototyping retrieval on your laptop, or want vectors in files with no server: Chroma.
- You want a hosted index with nothing to operate: Pinecone or turbopuffer.
Weaviate is an open-source database that stores data objects together with their vector embeddings, so the same collection can be searched by meaning, by keyword (BM25), or by both in one hybrid query. It is typically the retrieval store behind RAG, semantic search and agent features.
More about Weaviate
You define collections with typed properties and add objects to them. If you attach a vectorizer integration (OpenAI, Cohere, Voyage AI, Google, Ollama, Hugging Face and others), Weaviate sends the text to that model and stores the returned embeddings at import and query time; otherwise you supply your own vectors. Generative and reranker integrations let a single query retrieve objects and pass them to an LLM. You reach it over REST, gRPC or GraphQL, with official clients for Python, TypeScript/JavaScript, Go, Java and C#.
Useful for a small team: built-in multi-tenancy that puts each tenant on its own shard; filtered search that applies an allow-list from the inverted index inside the vector search; backups to S3, GCS or Azure Storage, including incremental ones; and role-based access control. You can run it with Docker, on Kubernetes, embedded from Python or JS, or on Weaviate Cloud (shared clusters on AWS or GCP, dedicated ones also on Azure).
Limits: Weaviate favours availability over consistency and has no transactions, so its docs suggest keeping data that needs strict serializability in a primary store. There is no native Windows build. The free cloud cluster allows one collection and 100,000 objects.
How Weaviate itself is built
6 tools, from its own code, website and Product Hunt page.
Weaviate pros and cons
What makers like most: Open source and free to run, with good docs and an active, friendly community. Their most common complaint: Search can be slow, with metadata filtering hurting vector query performance and hosted retrieval latency complaints.
Pros
- Open source and free to run, with good docs and an active, friendly community. HNweaviate.ioPHHN 2
- Scales to production workloads, including multi-tenant setups on its managed cloud. weaviate.ioPHPH 2weaviate.io 2
- Built-in vectorization and a GraphQL API let queries mix semantic search with structured data. PHHNHN 2
- Self-hosted costs stay low even with steady multi-user usage. HNHN 2
Themes that recur in makers' words and Hacker News comments, each linked to what it summarises, with how Product Hunt tags its reviews.
4.9★ from 13 reviews on Product Hunt ↗Reliability and open issues
The most wanted open issue on GitHub has 70 👍.
Most wanted on GitHub
- Reindex API (Inverted Index)👍 70 · opened Nov 2022 · active Feb 2026
- Possibility to update or/and delete a property👍 60 · opened Mar 2023 · active Apr 2026
- Feature request: Cross-tenant queries👍 27 · opened Sep 2023 · active Apr 2026
- [Proposal] Add [ImageRef] image reference type👍 19 · opened Mar 2023 · active Oct 2025
Who uses it
16 makers' products, each linked to the source that shows it, and 32 open-source projects that declare it in their code. The best known on Product Hunt: Cortex (3,553 followers), Zefi.ai (1,491 followers), Lamatic.ai (1,248 followers), Insight7 (943 followers), Quantera.ai (467 followers). 1 of the makers' products is gone now (domain lapsed, up for sale or taken down); it chose Weaviate while live.
“Lightning fast and personalised support for every query. They've helped us brainstorm some architecture level things.”
Vector DatabaseMaker says so · source ↗“Helped us scale on serverless infrastructure without worry.”
Vector DatabaseMaker says so · source ↗“We use Weaviate as a our vector database. Our main feed feature is powered by the vector embeddings Weaviate generates.”
Vector DatabaseMaker says so · source ↗“Weaviate gives us fast, semantic search across unstructured call data, crucial for surfacing insights in real time. Its native vector support and scalability made it the best fit for building an AI native platform like Insight7.”
Vector DatabaseMaker says so · source ↗“To store all our vector embeddings, now a staple for us to build forward. Their automatic 'load balancing' on which vectors are recently used is a game changer for system optimization”
Vector DatabaseMaker says so · source ↗“Unbody is built on top of Weaviate, making Unbody content API run 100% on a vector database. Weaviate modular architecture as well as user-friendly GraphQl API has played a vital role in our product.”
Vector DatabaseMaker says so · source ↗“Weaviate has let us manage our data and has supported us since the very beginning. Best team ever and amazing product”
Vector DatabaseMaker says so · source ↗“We chose Weaviate for its great developer experience and seamless integration with vector search and AI-powered queries. It makes working with embeddings and content indexing much simpler.”
Vector DatabaseMaker says so · source ↗None of these in makers' own words.
Open source: a project that declares Weaviate as a dependency in its public code — verifiable, but not necessarily a live product.
Alternatives to Weaviate
All alternatives by situation →Further reading
Questions makers ask about Weaviate
Is Weaviate good for production?
16 makers' products we track run on it, and 1 of them have since gone (the product, not necessarily because of Weaviate). The watch-out makers raise most: Search can be slow, with metadata filtering hurting vector query performance and hosted retrieval latency complaints.
What is Weaviate used for?
Open-source vector database combining vector, hybrid, and keyword search for AI and RAG applications. Makers' products we track use it for vector database (16).
What are the pros and cons of Weaviate?
Pros makers name: Open source and free to run, with good docs and an active, friendly community. Scales to production workloads, including multi-tenant setups on its managed cloud. Built-in vectorization and a GraphQL API let queries mix semantic search with structured data. Cons: Search can be slow, with metadata filtering hurting vector query performance and hosted retrieval latency complaints.
What are the alternatives to Weaviate?
For vector database, the other options makers' products we track chose most are Pinecone (53), Qdrant (18), Chroma (12), turbopuffer (8), LanceDB (3). Open source or self-hostable among them: Qdrant, Chroma, LanceDB. Each alternative, with the situation it fits, is on the Weaviate alternatives page.
How much does Weaviate cost per month?
At 10,000 users, Weaviate costs $253 a month on the cheapest plan that fits (Flex), from its pricing page.
When should you use Weaviate?
You need an open-source store with heavy filtering or hybrid search, self-hosted or managed (vector database). Something else fits better if: you already use Postgres and have up to a few million vectors, pick pgvector; you're prototyping retrieval on your laptop, or want vectors in files with no server, pick Chroma.
Can Weaviate create the embeddings for me?
Yes. With a model provider integration enabled, Weaviate sends an object's text properties to the provider (API-based ones like OpenAI or Cohere, or locally hosted ones like Ollama) and stores the vectors it returns. You can also import your own vectors instead. source ↗
Which languages have official clients?
Python, TypeScript/JavaScript, Go, Java and C#. Anything else can call the REST, gRPC or GraphQL APIs directly, and some community clients exist. source ↗
How do I keep each customer's data separate?
Enable multi-tenancy on the collection. Each tenant is stored on its own shard and its data is not visible to other tenants; you can also let Weaviate create tenants automatically on insert. source ↗
What are the limits of the free cloud cluster?
One free cluster per user, with up to 100,000 objects, one collection and three tenants. It is suspended after 7 days without activity and deleted after 30 days, with a warning email the day before. source ↗
Does Weaviate support transactions or strong consistency?
No. It has no transactions, each operation touches a single object, and a distributed cluster is eventually consistent. For data that needs strict serializability, the docs recommend a separate primary store with Weaviate alongside it. source ↗
Is Weaviate Cloud SOC 2 and HIPAA compliant?
Weaviate Cloud is SOC 2 Type II audited. HIPAA compliance is available only on the dedicated (Enterprise) deployment on AWS. source ↗
How do backups work when I self-host?
The backup feature writes a whole instance or selected collections to S3, GCS or Azure Storage with one command, supports incremental backups, and can restore to a different provider. There is no per-tenant restore; restoring brings back the whole collection. source ↗
Can I run it on Windows?
Only through Docker or WSL. There is no native Windows support, and the docs advise against Embedded Weaviate there. source ↗
Is Weaviate free?
Yes — there is a free tier a small product can run on; paid use starts at $45/mo. source ↗
Is Weaviate open source or self-hostable?
Open source, and you can self-host it. source ↗
Can AI coding agents work with Weaviate?
It serves an llms.txt docs index.
Who uses Weaviate?
16 makers' products we track, each with a source, and 32 open-source projects declare it in their code. The best known of them on Product Hunt: Cortex (3,553 followers), Zefi.ai (1,491 followers), Lamatic.ai (1,248 followers), Insight7 (943 followers), Quantera.ai (467 followers). source ↗