Integrate Jaguar Data with LlamaIndex
Retrieve documents from an existing Jaguar vector store into LlamaIndex.
Why it matters
Connect your persisted Jaguar data store to LlamaIndex for advanced data analysis and retrieval. Seamlessly ingest documents from Jaguar into LlamaIndex data structures.
Outcomes
What it gets done
Load documents from a Jaguar store into LlamaIndex.
Configure reader with Jaguar store details and API key.
Query Jaguar data with specified embeddings and metadata filters.
Prepare Jaguar data for downstream LlamaIndex applications.
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/li-reader-readers-jaguar | bash Overview
LlamaIndex Readers Integration: Jaguar
A LlamaIndex reader that retrieves vector-similarity matches from an existing Jaguar store via an explicit login/query/logout session. Use when you already store vector data in Jaguar and want a similarity-filtered subset loaded for retrieval.
What it does
The Jaguar Reader retrieves documents from an existing, already-persisted Jaguar vector store, so those documents can be used in a downstream LlamaIndex data structure. JaguarReader is initialized with a pod, store, vector_index, vector_type, vector_dimension, and endpoint url identifying the target Jaguar store.
Using it requires a Jaguar API key. The reader follows an explicit login/query/logout lifecycle: reader.login(jaguar_api_key=...) authenticates to the Jaguar server, load_data then takes an embedding vector, a k result count, a metadata_fields list, and a where query condition to fetch matching documents, and reader.logout() closes the session afterward.
When to use - and when NOT to
Use it when you already have vector data persisted in Jaguar and want a specific, vector-similarity-filtered subset of it loaded into LlamaIndex. Remember to call login before load_data and logout afterward - the reader manages an explicit session rather than authenticating per call. Do not use it as a general Jaguar administration tool; it is a read-only loader built around a vector query, not a way to write or manage the store.
Capabilities
load_data runs a vector-similarity query (an embedding, result count, metadata fields, and a where condition) against a Jaguar store and returns matching documents, within an explicit login/logout session.
How to install
pip install llama-index-readers-jaguar
Requires a Jaguar API key and an already-persisted Jaguar store (pod, store, vector index, vector type, and dimension).
Who it's for
Developers who already store vector data in Jaguar and want a vector-similarity-filtered subset loaded into LlamaIndex for retrieval or question-answering.
Source README
LlamaIndex Readers Integration: Jaguar
Overview
Jaguar Reader retrieves documents from an existing persisted Jaguar store. These documents can then be used in a downstream LlamaIndex data structure.
Installation
You can install Jaguar Reader via pip:
pip install llama-index-readers-jaguar
To use Jaguar Reader, you must have an API key. Here are the installation instructions
Usage
from llama_index.readers.jaguar import JaguarReader
### Initialize JaguarReader
reader = JaguarReader(
pod="<Pod Name>",
store="<Store Name>",
vector_index="<Vector Index Name>",
vector_type="<Vector Type>",
vector_dimension="<Vector Dimension>",
url="<Endpoint URL>",
)
### Login to Jaguar server
reader.login(jaguar_api_key="<Jaguar API Key>")
### Load data from Jaguar
documents = reader.load_data(
embedding="<Embedding Vector>",
k=10,
metadata_fields=["<Metadata Field 1>", "<Metadata Field 2>"],
where="<Query Condition>",
)
### Logout from Jaguar server
reader.logout()
This loader is designed to be used as a way to load data into
LlamaIndex.
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.