API Reference¶
This section provides detailed API documentation for EmbedCache.
Overview¶
EmbedCache provides:
- REST API - HTTP endpoints for embedding generation and URL processing
- Rust API - Library interface for direct integration
REST API¶
The REST API is available when running EmbedCache as a service.
Base URL: http://localhost:8081 (default)
Endpoints¶
| Method | Endpoint | Description |
|---|---|---|
| POST | /v1/embed |
Generate embeddings for text |
| POST | /v1/process |
Process URL and generate embeddings |
| GET | /v1/params |
List supported features |
Interactive Documentation¶
When the service is running, interactive API documentation is available:
- Swagger UI - Interactive API explorer
- ReDoc - Clean API documentation
- RapiDoc - Modern API documentation
- Scalar - Beautiful API reference
- OpenAPI JSON - Raw OpenAPI specification
Rust API¶
The library API provides direct access to EmbedCache functionality.
Main Types¶
| Type | Description |
|---|---|
FastEmbedder |
Primary embedding generator |
WordChunker |
Word-based text chunking |
LLMConceptChunker |
LLM concept-based chunking |
LLMIntrospectionChunker |
LLM introspection chunking |
AppState |
Application state container |
Config |
Request configuration |
Main Functions¶
| Function | Description |
|---|---|
initialize_db_pool |
Create database connection pool |
initialize_models |
Load embedding models |
initialize_chunkers |
Create chunker instances |
get_embedding_model |
Get model by name |
generate_hash |
Generate cache hash |
fetch_content |
Fetch URL content |
Response Codes¶
| Code | Description |
|---|---|
| 200 | Success |
| 400 | Bad Request (invalid model/chunker) |
| 500 | Internal Server Error |