Expand description
§Vectorless
A reasoning-native document engine for AI.
It will reason through any of your structured documents — PDFs, Markdown, reports, contracts — and retrieve only what’s relevant. Nothing more, nothing less.
§Quick Start
use vectorless::{EngineBuilder, IndexContext, QueryContext};
#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
let engine = EngineBuilder::new()
.with_key("sk-...")
.with_model("gpt-4o")
.with_endpoint("https://api.openai.com/v1")
.build()
.await?;
let result = engine.index(IndexContext::from_path("./document.md")).await?;
let doc_id = result.doc_id().unwrap();
let result = engine.query(
QueryContext::new("What is this about?")
.with_doc_ids(vec![doc_id.to_string()]),
).await?;
println!("{}", result.content);
Ok(())
}Structs§
- Config
- Main configuration for vectorless.
- Document
Graph - A workspace-scoped document relationship graph.
- Document
Graph Node - A document node in the graph.
- Document
Info - Document info for listing.
- Document
Structure - Document structure for JSON export.
- Document
Tree - A hierarchical document tree structure.
- Edge
Evidence - Evidence for why two documents are connected.
- Engine
- The main Engine client.
- Engine
Builder - Builder for creating a
Engineclient. - Event
Emitter - Event emitter for client operations.
- Evidence
Item - A single piece of evidence with source attribution.
- Failed
Item - A failed item in a batch operation.
- Graph
Edge - An edge connecting two documents.
- Index
Context - Context for document indexing operations.
- Index
Item - A single indexed document item.
- Index
Metrics - Performance metrics for the indexing pipeline.
- Index
Options - Options for indexing a document.
- Index
Result - Result of a document indexing operation.
- LlmMetrics
Report - LLM metrics report.
- Metrics
Report - Complete metrics report.
- NodeId
- Unique identifier for a node in the document tree.
- Query
Context - Context for a query operation.
- Query
Metrics - Query execution metrics.
- Query
Result - Result of a document query.
- Query
Result Item - A single document’s query result.
- Reasoning
Index Config - Configuration for building and using the reasoning index.
- Retrieval
Metrics Report - Retrieval metrics report.
- Structure
Node - A node in the document structure for JSON export.
- TocConfig
- Configuration for ToC generation.
- TocEntry
- A flat ToC entry.
- TocNode
- A node in the Table of Contents.
- TocView
- ToC view generator.
- Tree
Node - A node in the Vectorless document tree.
- Weighted
Keyword - A keyword with its aggregate weight across the document.
Enums§
- Build
Error - Error during client build.
- Document
Format - Supported document formats.
- Error
- The main error type for vectorless operations.
- Index
Event - Indexing operation events.
- Index
Mode - Document indexing behavior mode.
- Query
Event - Query operation events.
- Workspace
Event - Workspace operation events.
Type Aliases§
- Confidence
- Confidence score of the query result (0.0–1.0).
- Result
- A specialized result type for vectorless operations.