ai/studio

Quantum Minds Operator Categories

Introduction

Operators are the fundamental building blocks of Quantum Minds. Each operator performs a specific function, taking inputs and producing outputs that can be connected to other operators. This document provides an overview of the operator categories available in Quantum Minds.

Operator Classification

Operators in Quantum Minds are organized into the following categories:

Category	Description	Common Uses
SQL	Database operations and SQL query generation	Database analysis, data extraction, schema understanding
Table	Tabular data manipulation and analysis	Data transformation, visualization, summarization
Document	Processing PDFs and unstructured text	Document understanding, RAG operations, knowledge extraction
MongoDB	NoSQL database operations	Document database querying, flexible data model analysis
LLM	Integration with external language models	Natural language processing, content generation
Excel	Spreadsheet operations	Data import/export, report generation, presentation creation
Media	Processing images, audio, and video	Content generation, speech processing, multimedia analysis
API	Working with external services	API execution, endpoint integration, service connections
Code	Running and generating code	Custom logic, algorithmic processing, code synthesis
Flow	Controlling the flow of operations	Conditional branching, decision logic
Vector	Vector embedding and operations	Semantic search, similarity analysis, vector database interaction
Finance	Financial data processing	Market data analysis, financial reporting, investment analysis
ML	Machine learning operations	Predictive modeling, classification, clustering, forecasting
UI	User interface component generation	Dashboard elements, cards, interactive displays
Monitor	System monitoring operations	Performance tracking, observability, log analysis

Model Integration

Many operators in Quantum Minds can utilize various AI models to perform their functions. When configuring these operators, you can select from supported models including:

Fireworks models (llama-v3p1-70b-instruct, mixtral-8x22b-instruct, etc.)
Groq models (llama3-70b-8192, gemma2-9b-it, etc.)
OpenAI models (gpt-4o, gpt-3.5-turbo, etc.)
Anthropic models (claude-3-sonnet, claude-3-5-sonnet, etc.)
Mistral models (mistral-7b, mixtral-8x22b, etc.)
Specialized models for specific tasks (whisper, DALL-E, etc.)

Input and Output Patterns

Operators follow consistent patterns for their inputs and outputs:

Common Input Types

prompt: Text instructions for what the operator should do
dataframe: Tabular data as input
file: File content for processing
dataset: Reference to a data source
collection: Reference to a knowledge collection
trigger: Control signal that initiates the operator

Common Output Types

content: The primary result of the operator
type: The format of the output (markdown, table, recharts, etc.)
dataframe: Tabular data output
graph: Visualization data
error: Error information if the operation fails

Operator Versioning

Many operators in Quantum Minds have multiple versions that offer enhanced capabilities or different approaches. Versioned operators are indicated by a "V" followed by a number (e.g., TextToSQLV4, TableToGraphV3).

When selecting operators for your mind, consider:

Using the latest version for the most advanced capabilities
Using specific versions if you need particular features or compatibility
Checking the documentation for each version to understand differences

Categories in Detail

SQL Operators

SQL operators enable interaction with relational databases, allowing natural language querying, SQL generation, and data extraction. They are ideal for working with structured data in traditional databases.

Table Operators

Table operators focus on manipulating, analyzing, and visualizing tabular data. These are essential for data transformation, cleaning, and preparing data for insights.

Document Operators

Document operators process unstructured and semi-structured text data, including PDFs, web content, and raw text. They enable document understanding and knowledge extraction.

MongoDB Operators

MongoDB operators provide specialized functionality for working with NoSQL databases, handling flexible data models and document-oriented storage.

LLM Operators

LLM operators facilitate interaction with external large language models, allowing your minds to leverage the capabilities of various AI providers.

Excel Operators

Excel operators handle spreadsheet data, enabling import/export and specialized processing of Excel files, including transformation to presentations.

Media Operators

Media operators process non-text content including images, audio, and video, enabling multimodal AI applications.

API Operators

API operators enable interaction with external services and systems through API calls, extending the capabilities of your minds beyond the platform.

Code Operators

Code operators allow execution and generation of code, primarily Python, to perform custom logic and algorithms within your minds.

Flow Operators

Flow operators control the execution path of your mind, enabling conditional logic and branching based on specific criteria.

Vector Operators

Vector operators handle embedding generation and interaction with vector databases, enabling semantic search and similarity-based operations.

Finance Operators

Finance operators provide specialized functionality for financial data analysis, particularly integrating with external financial data sources.

ML Operators

ML operators facilitate machine learning tasks including model selection, training, evaluation, and prediction across various ML paradigms.

Next Steps

Explore the detailed documentation for each operator category to learn about the specific operators available, their inputs, outputs, and best practices for use.

Overview | SQL Operators | Table Operators | Document Operators