> For the complete documentation index, see [llms.txt](https://vibemx.gitbook.io/neuronai-studio/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://vibemx.gitbook.io/neuronai-studio/knowledge-bases/vector-stores.md).

# Vector Stores

NeuronAI Studio registers first-party [Neuron AI vector stores](https://docs.neuron-ai.dev/rag/vector-store) in `VectorStoreFactory`. Each knowledge base picks a driver and stores connection fields in `vector_store_config`.

Secrets should use **env var names** (`key_env` / `api_key_env`), not raw API keys in the database.

## Driver matrix

| Driver          | Extra Composer package              | Main config fields                                                                  |
| --------------- | ----------------------------------- | ----------------------------------------------------------------------------------- |
| `file`          | —                                   | optional `directory` (defaults to `rag.storage_path`)                               |
| `memory`        | —                                   | none (volatile; good for tests)                                                     |
| `pinecone`      | —                                   | `key_env`, `index_url`, `namespace`                                                 |
| `qdrant`        | —                                   | `collection_url`, `key_env`, `dimension`                                            |
| `chroma`        | —                                   | `collection`, `host`, `key_env`, `tenant`, `database`                               |
| `weaviate`      | —                                   | `collection`, `host`, `key_env`                                                     |
| `meilisearch`   | —                                   | `index_uid`, `host`, `key_env`, `embedder`, `dimension`                             |
| `mariadb`       | — (MariaDB ≥11.7)                   | `connection`, `table`                                                               |
| `elasticsearch` | `elasticsearch/elasticsearch`       | `hosts`, `api_key_env`, `index`                                                     |
| `opensearch`    | `opensearch-project/opensearch-php` | `base_uri`, `index`                                                                 |
| `typesense`     | `typesense/typesense-php`           | `api_key_env`, nodes via `host`/`port`/`protocol`, `collection`, `vector_dimension` |
| `phpvector`     | `neuron-core/php-vector`            | `path`                                                                              |

Optional packages are listed under `suggest` in the Studio `composer.json`. Selecting a driver without its client installed fails with a `composer require …` hint.

## MariaDB setup

Create a VECTOR table (adjust dimension to match your embedder):

```sql
CREATE TABLE IF NOT EXISTS rag_documents (
    id UUID NOT NULL PRIMARY KEY,
    content TEXT,
    sourceType VARCHAR(255),
    sourceName VARCHAR(255),
    metadata JSON,
    embedding VECTOR(1536) NOT NULL,
    VECTOR INDEX (embedding)
);
```

Point the knowledge base at your Laravel DB connection name and table.

## Custom drivers

Override or add drivers at boot:

```php
use DigitalElvis\NeuronAIStudio\Runtime\Rag\VectorStoreFactory;

app(VectorStoreFactory::class)->extend('my-store', function ($knowledgeBase, array $options) {
    return new \App\Neuron\MyVectorStore(/* ... */);
});
```

Then add an entry under `config('neuronai-studio.rag.vector_stores')` so it appears in the UI.

## Env reference

| Key                        | Env                                  | Default                               |
| -------------------------- | ------------------------------------ | ------------------------------------- |
| `rag.default_vector_store` | `NEURONAI_STUDIO_RAG_VECTOR_STORE`   | `file`                                |
| `rag.storage_path`         | `NEURONAI_STUDIO_RAG_STORAGE_PATH`   | `storage/app/neuronai-studio/rag`     |
| `rag.documents_disk`       | `NEURONAI_STUDIO_RAG_DOCUMENTS_DISK` | `local`                               |
| `rag.documents_path`       | `NEURONAI_STUDIO_RAG_DOCUMENTS_PATH` | `neuronai-studio/knowledge-documents` |
| `rag.async_ingest`         | `NEURONAI_STUDIO_RAG_ASYNC_INGEST`   | `true`                                |

Full table: [Configuration — RAG](/neuronai-studio/reference/configuration.md#rag).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://vibemx.gitbook.io/neuronai-studio/knowledge-bases/vector-stores.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.