System Config

Manage LLM providers, API keys & workbench parameters.

Doc Ingestion

Import, parse & extract clean text from multi-format files.

Chunking Lab

Benchmark 6 chunkers: fixed, recursive, semantic, markdown, sentence & code-aware.

Embedding Studio

Explore & inspect high-dimensional 3D vector spaces.

Pipeline Builder

Compose dense, sparse BM25 & hybrid pipelines.

RAG Execution

Stream live RAG queries with real-time telemetry.

RAG Evaluator

Quantify context relevance, faithfulness & accuracy.

Document Ingestion & Parsing

Import, parse & extract clean text from multi-format files in the workspace.

Step 1

Select File

Loading Workspace Files...

Step 2

Review & Ingest

| Select a file to inspect

Select a file from the registry to view data.

Chunking Laboratory

Select an ingested file, benchmark 6 chunkers, and visualize splits dynamically.

Workspace Files

Loading Files...

Chunking Strategy

Document Workspace

| Select a file

First, select an ingested file from the Workspace.
Then select a strategy to start chunking.

3D Vector Embedding Studio

Vectorize chunked text using multiple models and visualize semantic space.

1 Select Chunked File

2 Select Strategies

Showing only computed strategies for the selected file.

Select a file first

3 Select Embedders

Execution Log

Embedded Registry

Select a file to view its combinations

Loading registry...

Embedding Visualization

Select a combination from the registry to view.

VISUALIZATION READY

Select a combination to see details and render 3D vectors.

Node Details

0.000

Content

Metadata

Doc:--

Strat:--

DB ID:--

Pipeline Architecture Builder

Compose dense, sparse BM25 & hybrid retrieval pipelines.

LLM Provider & Model are set globally by environment, not in pipelines.

Step 1

Create Pipeline

Pipeline Name

Chunker Module

Embedder Module

Retriever Module

Reranker Module (Optional)

Step 2

Registered Pipelines

CUSTOM PIPELINES

PRESET PIPELINES

RAG Execution Engine

Real-time Inference & Context Verification

1 Pipeline

2 Document

Step 3

Pipeline Dependency Verification

NOT READY

Ingestion

Parsed text

Chunking

Req: ?

Embedding

Req: ?

Step 4 Execute Query

Awaiting Query Execution

Retrieved Context

Source chunks passed to the LLM context window

No context loaded

System & Provider Configuration

Provision workspace structure and configure default LLM generation models.

Workspace Provisioning

Storage Architecture Matrix

Loading Matrix...

LLM Infrastructure Engines

Click card to focus; set active provider when key is configured

API Security & Endpoint Vault

Configuring

Awaiting Vault Access...

Confirm Chunk Execution

System Config

Doc Ingestion

Chunking Lab

Embedding Studio

Pipeline Builder

RAG Execution

RAG Evaluator

Document Ingestion & Parsing

Select File

Review & Ingest

Chunking Laboratory

Workspace Files

Chunking Strategy

Document Workspace

3D Vector Embedding Studio

1 Select Chunked File

2 Select Strategies

3 Select Embedders

Embedded Registry

Embedding Visualization

VISUALIZATION READY

Node Details

Pipeline Architecture Builder

Create Pipeline

Registered Pipelines

CUSTOM PIPELINES

PRESET PIPELINES

RAG Execution Engine

Pipeline Dependency Verification

Retrieved Context

RAG Evaluator

Execution Details

RAG Arena

Triad Radar Footprint

Metric Breakdown

Arena Leaderboard

System & Provider Configuration

Workspace Provisioning

Storage Architecture Matrix

LLM Infrastructure Engines

API Security & Endpoint Vault