API Reference
Comprehensive API documentation for RedisVL, generated from source code comments.
|
The API documentation is automatically generated from the latest release and includes:
|
Core Module
The RedisVL core module provides comprehensive functionality for vector search, semantic caching, and AI-native Redis applications.
Key Packages
-
Index:
com.redis.vl.index- Search index creation and management -
Query:
com.redis.vl.query- Vector and hybrid query builders -
Schema:
com.redis.vl.schema- Index schema definition and validation -
Cache:
com.redis.vl.extensions.cache- Semantic and embeddings caching -
Vectorizers:
com.redis.vl.utils.vectorize- Embedding generation with LangChain4J and ONNX -
Rerankers:
com.redis.vl.utils.rerank- Document reranking with cross-encoder models
Complete API Reference
For a unified view of all packages and their interactions:
This aggregated documentation provides:
-
Complete inheritance hierarchies across packages
-
Unified search across all classes
-
Package overview documentation for the entire library
Most Important Classes
Core Classes
| Class | Description |
|---|---|
|
Main class for creating and managing vector search indexes |
|
Schema definition for indexes, supports YAML and programmatic configuration |
|
Builder for vector similarity queries with filtering |
|
Query filter builder for hybrid search (tag, numeric, text filters) |
Caching Classes
| Class | Description |
|---|---|
|
LLM response caching with semantic similarity matching |
|
Cache for storing and retrieving embeddings to avoid recomputation |
|
Result object containing cached data and metadata |
Vectorizer Classes
| Class | Description |
|---|---|
|
Vectorizer using LangChain4J embedding models (OpenAI, Azure, Ollama, etc.) |
|
Local ONNX-based vectorizer for Sentence Transformers models |
|
Abstract base class for custom vectorizer implementations |
|
Utility for downloading ONNX models from Hugging Face |
Reranker Classes
| Class | Description |
|---|---|
|
Reranker using HuggingFace cross-encoder models via ONNX Runtime |
|
Abstract base class for reranker implementations |
|
Result object containing reranked documents and optional relevance scores |
|
ONNX model loader for cross-encoder models with tokenization support |
Integration with IDE
For the best development experience, add the Javadoc JAR to your IDE:
IntelliJ IDEA
-
Go to File → Project Structure → Libraries
-
Select the RedisVL library
-
Click + and add the Javadoc JAR
-
Apply changes