---
title: Roadmap
description: The main features that we are currently working on
canonical: https://www.paradedb.com/docs/project/roadmap
---

## In Progress

### Vector Search

- **[Native vector search (beta)](/reference/vector/overview)**. Search pgvector's `vector` type with the ParadeDB index, combining vector similarity with metadata filters and full-text search in a single query.
- **Incremental index maintenance**. Vector search quality currently degrades as indexed rows are updated. We are implementing SPFresh to maintain index quality incrementally.
- **Remaining pgvector types**. Implement support for pgvector's `halfvec`, `sparsevec`, and `bit` types.

### JOIN Improvements

- **[Accelerated JOINs (beta)](/reference/joins/overview)**. Apply search filters and aggregates directly in the index to speed up queries that join tables.
- **Smarter JOIN planning for search indexes**. Apply index-aware optimizations and cost estimation strategies when multiple ParadeDB-indexed tables are joined.

### Ecosystem Integrations

- **ORMs**. Official support for Prisma, Diesel, and others is coming. [Drizzle](https://github.com/paradedb/drizzle-paradedb), [Django](https://github.com/paradedb/django-paradedb), [SQLAlchemy](https://github.com/paradedb/sqlalchemy-paradedb), [Rails](https://github.com/paradedb/rails-paradedb), and [Entity Framework Core](https://github.com/paradedb/efcore-paradedb) are already available.
- **AI Frameworks**. Official support for LangChain, LlamaIndex, CrewAI, and others is coming.

### Cloud Platform

- **Distributed search**. Scale search across multiple nodes to support larger datasets and workloads.
- **Reranking**. Improve search relevance by reranking retrieved results with dedicated ranking models.

## Planned

### Object Storage

- **Offload index segments to object storage**. Reduce storage costs for large datasets by storing index segments in object storage.

### Deeper Analytics Improvements

- **Push Postgres visibility rules into the index**. This is currently a filter applied post index scan that adds overhead to large scans.

---

## We're Hiring

We're tackling some of the hardest and (in our opinion) most impactful problems in Postgres. If you want to be a part of it,
check out our [open roles](https://paradedb.notion.site)!