RushDB
Instant graph memory for AI agents.
SubscriptionAI Agents2,215 impressions#8 of its week2 comments
Comments
>log in to comment- Zain Sheikh· 4mo ago
Pairing vector search with the graph layer is the interesting part for retrieval - agents can jump from a semantic match to connected context in one hop. Is the vector index built in, or do you bring your own embeddings?
- PDFops· 4mo ago
The no-schema "push any JSON, get relationship-aware records" is the part that sells it — schema migrations are exactly what kills using a real graph DB as agent memory. The thing I'd want to know before wiring it into an agent loop: across many writes of the same entity in slightly different shapes, does it dedup/merge into one node, or does each push create new ones? Unbounded node growth is what quietly turns agent memory into agent landfill. Built on Neo4j is reassuring either way — starring it.
RushDB provides a graph-plus-vector database and memory layer that lets AI agents store, query and relate JSON data without schema or migrations.
- for
- Developers building AI agents, applications and analytics that need connected graph and vector search.
- pricing
- freemium
- license
- open source
- @rushdb/mcp-server@2.13.02 months ago
- @rushdb/skills@2.12.02 months ago
- @rushdb/agent-memory-contract@0.2.02 months ago
Key features
- Schema-less JSON import — Push nested JSON and RushDB auto-creates linked graph records, live schema and embeddings.
- Unified query API — Serve application reads, natural-language requests and analytics from the same model via REST, SDK or MCP.
- Live schema inspection — Applications can read the generated schema, fields and relationships before building filters.
- Semantic search & vector embeddings — Records are enriched with embeddings enabling meaning-based retrieval.
- Multi-deployment options — Run as managed cloud, self-hosted Docker, or embed in your product with the same API.
- Usage-based pricing — Standard reads are free; writes consume Knowledge Units (KUs) with tiered plans.
Use cases
- Persist context across multiple AI agent runs for a customer support bot
- Enable semantic search over operational data without a separate vector store
- Run analytics on connected business entities using a single graph model
RushDB pricing
- Free$0100K KU/month · 2 projects · Full REST API & SDKs · Standard reads free
- Start$8/month250K KU/month · 1 team member · Self-hosted & External DB · Community support
RushDB vs alternatives
 RushDB |  Fine |  SuperAGI Cloud |  Pocketenv |  GraphBit | |
|---|---|---|---|---|---|
| Best for | Graph-aware memory for AI agents | AI agent development platform | Managed autonomous agent hosting | Agent sandbox environment | Agentic AI framework |
| Pricing | Freemium | Free | Free | Free | Free |
| DevHunt upvotes | 3 | 75 | 43 | 73 | 16 |
| Launched | Jun 2026 | Jan 2023 | Jan 2023 | Apr 2026 | Oct 2025 |
- RushDB vs Fine: Fine focuses on building AI agents, while RushDB provides the underlying graph-vector memory layer.
- RushDB vs SuperAGI Cloud: SuperAGI Cloud runs autonomous agents, whereas RushDB supplies the persistent graph memory they can use.
- RushDB vs Pocketenv: Pocketenv offers a sandbox runtime for agents; RushDB adds a persistent, searchable graph memory layer.
- RushDB vs GraphBit: GraphBit is a Rust-first agentic framework; RushDB is a language-agnostic graph-vector database for agents.
RushDB FAQ
Is RushDB a database or a layer over my existing database?+
Both; it can be the primary backend or a shared layer that syncs with PostgreSQL, MongoDB, object storage, etc.
Do I need to replace PostgreSQL, MongoDB, or my warehouse?+
No, you can keep existing systems and use RushDB as a shared context layer.
Can RushDB serve production application traffic?+
Yes, the Production plan provides managed usage and predictable overages for production workloads.
What happens when I hit my KU limit on the Free plan?+
Writes pause until the next billing period; reads and standard queries continue, with email alerts at usage thresholds.
What is a Knowledge Unit (KU)?+
A KU measures the records, properties, relationships and embeddings created by a write operation.
Summarized by DevHunt from rushdb.com · Sep 28, 2026. Details may change; check the official site.
.png?width=64){kind=logo}
.svg?width=64)
