llama-index-vector-stores-velesdb 1.8.0

LlamaIndex VectorStore for VelesDB: The Local AI Memory Database. Microsecond RAG retrieval.

LlamaIndex VelesDB Integration

VelesDB vector store integration for LlamaIndex.

Features

• 🚀 Sub-millisecond search — SIMD-optimized vector retrieval
• 📦 Self-contained — Single VelesDB binary, no external services required
• 🔒 Local-first — All data stays on your machine
• 🧠 RAG-ready — Built for Retrieval-Augmented Generation
• 🔀 Multi-Query Fusion — Native MQG support with RRF/Weighted strategies

Installation

pip install llama-index-vector-stores-velesdb

Quick Start

from llama_index.core import VectorStoreIndex, SimpleDirectoryReader
from llamaindex_velesdb import VelesDBVectorStore

# Create vector store
vector_store = VelesDBVectorStore(
    path="./velesdb_data",
    collection_name="my_docs",
    metric="cosine",
)

# Load and index documents
documents = SimpleDirectoryReader("./data").load_data()
index = VectorStoreIndex.from_documents(
    documents,
    vector_store=vector_store,
)

# Query
query_engine = index.as_query_engine()
response = query_engine.query("What is VelesDB?")
print(response)

Usage with Existing Index

from llama_index.core import VectorStoreIndex
from llamaindex_velesdb import VelesDBVectorStore

# Connect to existing data
vector_store = VelesDBVectorStore(path="./existing_data")
index = VectorStoreIndex.from_vector_store(vector_store)

# Query
query_engine = index.as_query_engine()
response = query_engine.query("Summarize the key points")

API Reference

VelesDBVectorStore

VelesDBVectorStore(
    path: str = "./velesdb_data",      # Database directory
    collection_name: str = "llamaindex", # Collection name
    metric: str = "cosine",             # Distance metric
    storage_mode: str = "full",         # Storage / quantization mode
)

Parameters:

Parameter	Type	Default	Description
path	str	"./velesdb_data"	Path to database directory
collection_name	str	"llamaindex"	Name of the collection
metric	str	"cosine"	Distance metric: cosine, euclidean, dot, hamming, jaccard
storage_mode	str	"full"	Storage mode: full (f32), sq8 (4× compression), binary (32× compression)

Methods:

Method	Description
Core Operations
add(nodes)	Add nodes with embeddings
add_bulk(nodes)	Bulk insert (2-3x faster for large batches)
delete(ref_doc_id)	Delete by document ID
get_nodes(node_ids)	Retrieve nodes by their IDs
flush()	Flush pending changes to disk
Search
query(query)	Query with vector
batch_query(queries)	Batch query multiple vectors in parallel
multi_query_search(query_embeddings, ...)	Multi-query fusion search ⭐ NEW
hybrid_query(query_str, query_embedding, ...)	Hybrid vector+BM25 search
text_query(query_str, ...)	Full-text BM25 search
velesql(query_str, params)	Execute VelesQL query
Utilities
get_collection_info()	Get collection metadata
is_empty()	Check if collection is empty

Advanced Features

Multi-Query Fusion (MQG)

Search with multiple query embeddings and fuse results using various strategies. Perfect for RAG pipelines using Multiple Query Generation (MQG).

from llamaindex_velesdb import VelesDBVectorStore

vector_store = VelesDBVectorStore(path="./velesdb_data")

# Basic usage with RRF (Reciprocal Rank Fusion)
results = vector_store.multi_query_search(
    query_embeddings=[emb1, emb2, emb3],  # Multiple query reformulations
    similarity_top_k=10,
    fusion="rrf",
    fusion_params={"k": 60}
)

# With weighted fusion (like SearchXP's scoring)
results = vector_store.multi_query_search(
    query_embeddings=[emb1, emb2],
    similarity_top_k=10,
    fusion="weighted",
    fusion_params={
        "avg_weight": 0.6,   # Average score weight
        "max_weight": 0.3,   # Maximum score weight  
        "hit_weight": 0.1,   # Hit ratio weight
    }
)

for node in results.nodes:
    print(f"{node.metadata}: {node.text[:50]}...")

Fusion Strategies:

• "rrf" - Reciprocal Rank Fusion (default, robust to score scale differences)
• "average" - Mean score across all queries
• "maximum" - Maximum score from any query
• "weighted" - Custom combination of avg, max, and hit ratio

Hybrid Search (Vector + BM25)

from llamaindex_velesdb import VelesDBVectorStore

vector_store = VelesDBVectorStore(path="./velesdb_data")

# Hybrid search combining semantic and keyword matching
results = vector_store.hybrid_query(
    query_str="machine learning optimization",
    query_embedding=embedding_model.get_query_embedding("machine learning optimization"),
    similarity_top_k=10,
    vector_weight=0.7  # 70% vector, 30% BM25
)
for node in results.nodes:
    print(node.text)

Full-Text Search (BM25)

# Pure keyword-based search without embeddings
results = vector_store.text_query(
    query_str="VelesDB performance",
    similarity_top_k=5
)

Performance

Measured on CI runners (ubuntu-latest, 2-core). Local hardware will be faster. Source: benchmarks/baseline.json (v1.7.0).

Operation	Latency	Notes
Search (10K / 128D, k=10)	~0.34 ms	HNSW + SIMD cosine
Hybrid (vector + filter)	~0.27 ms	Filtered vector search
Batch insert (10K / 128D)	~9 s total	Sequential HNSW build

Comparison with Other Stores

Feature	VelesDB	Chroma	Pinecone
Deployment	Local binary	Docker	Cloud
Cost	Free	Free	$$$
Offline	✅	✅	❌

License

MIT License (this integration)

VelesDB Core and Server are licensed under VelesDB Core License 1.0 (source-available). See the root LICENSE for details.