Skip to content

Comprehensive data capture and relational database for OpenRouter AI models

License

Notifications You must be signed in to change notification settings

CreativeSystemsDevelopment/Open-Model

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Open-Model

A comprehensive data capture and relational database for OpenRouter AI models.

Overview

Open-Model is a data-focused project that captures complete model information from OpenRouter's API and structures it into a well-designed relational database. This repository contains only the data collection, processing, and database schema - no application interface.

Purpose

This project serves as the foundational data layer for AI model analysis and research. It provides:

  • Complete OpenRouter Model Data: All 326+ models with full metadata
  • Structured Relational Database: Proper entity relationships and constraints
  • Data Export Capabilities: Multiple formats for development and analysis
  • Raw Data Preservation: Complete original API responses for full fidelity

Project Scope

This repository is data-only. It focuses exclusively on:

  • Data capture from OpenRouter API
  • Relational database schema design
  • Data validation and quality assurance
  • Export utilities for downstream use

Applications, visualizations, and interfaces are built separately using this data foundation.

Features

  • Complete Data Capture: All 326+ OpenRouter models with full metadata
  • Relational Design: Proper entity relationships, foreign keys, and constraints
  • Provider Management: Track model providers and their relationships
  • Capability Classification: Organize models by capabilities and use cases
  • Pricing Analysis: Comprehensive pricing structure with micro-precision
  • Technical Specifications: Architecture details, context lengths, and parameters
  • Raw Data Preservation: Complete original JSON for full fidelity
  • Export Capabilities: Multiple output formats for development use

Architecture

Database Design Philosophy

  • Relational Foundation: Rock-solid PostgreSQL schema with proper relationships
  • Data Integrity: Foreign key constraints and validation rules
  • Comprehensive Capture: All fields preserved, including null/experimental ones
  • Future-Proof: Extensible design for new OpenRouter features

Core Entities

  • Providers: Model providers (OpenAI, Anthropic, Google, etc.)
  • Model Families: Logical groupings (GPT-4, Claude-3, Llama-2)
  • Models: Individual model instances with complete metadata
  • Capabilities: Structured capability classification system
  • Pricing: Multi-dimensional pricing with micro-precision
  • Technical Specs: Architecture details and parameters

Prerequisites

  • Rust 1.70+
  • PostgreSQL 14+
  • OpenRouter API access

Installation

git clone https://github.com/[username]/openrouter-model-database.git
cd openrouter-model-database
cargo build --release

Usage

1. Database Setup

# Create PostgreSQL database
createdb openrouter_models

# Run migrations
cargo run -- migrate

2. Data Capture

# Fetch and process all OpenRouter models
cargo run -- capture --api-key your_openrouter_key

# Generate reports
cargo run -- analyze

3. Export Data

# Export to various formats
cargo run -- export --format json
cargo run -- export --format csv
cargo run -- export --format sql

Data Quality

  • 326+ models captured and analyzed
  • 100% coverage for essential fields
  • Complete preservation of all API data
  • Validation rules ensure data integrity
  • Change detection with SHA256 hashing

Export Formats

  • PostgreSQL Dump: Complete database with schema and data
  • JSON: Structured data for development use
  • CSV: Spreadsheet-compatible format for analysis
  • Raw JSON: Original OpenRouter API responses

Contributing

This project focuses specifically on data capture and relational database creation. Contributions should maintain this scope and ensure data quality and integrity.

License

MIT License - see LICENSE file for details

Roadmap

  • Complete data capture implementation
  • Comprehensive database schema
  • Data validation and quality assurance
  • Export functionality
  • Documentation and examples
  • Future: Semantic search layer (separate phase)
  • Future: 3D visualization interface (separate phase)

Technical Details

Database Schema

  • Fully normalized relational design
  • Foreign key constraints for data integrity
  • Optimized indexes for query performance
  • Extensible structure for future OpenRouter features

Data Processing

  • Real-time API data fetching
  • Intelligent capability extraction
  • Provider relationship mapping
  • Pricing tier classification
  • Technical specification parsing

About

Comprehensive data capture and relational database for OpenRouter AI models

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages