Open Router Model Scraper

🤖 Open Router Model Scraper is a powerful Apify Actor designed to discover and analyze comprehensive Open Router Model data from the OpenRouter API. This tool provides detailed Open Router Model information including pricing, capabilities, context lengths, and architectural details. Whether you're researching AI models, comparing Open Router Model costs, or building AI applications, the Open Router Model Scraper delivers detailed Open Router Model intelligence efficiently.

With direct OpenRouter API integration, multi-keyword filtering, comprehensive field extraction, and pricing calculation, the Open Router Model Scraper ensures complete discovery of relevant Open Router Model options. It focuses on key Open Router Model metrics including pricing, context length, vision support, tool calling, and JSON output capabilities, making it an essential tool for Open Router Model research and AI application development.

---

📋 Table of Contents

• Features
• Data Source
• How It Works
• Input
• Output
• Technical Stack
• Data Fields
• Model Capabilities
• Use Cases
• Quick Start
• Configuration
• Performance
• Important Notes
• Keywords
• Changelog
• Support

---

🔥 Features

• OpenRouter API Integration – Direct integration with OpenRouter to access 200+ Open Router Model options.
• Multi-Keyword Search – Filter Open Router Model by provider, name, or capabilities using multiple keywords (up to 10).
• Comprehensive Model Data – Extracts 20+ fields including pricing, context length, and capabilities.
• Pricing Calculation – Converts pricing from API format to human-readable per-million-token rates.
• Vision Support Detection – Identifies Open Router Model with image/vision capabilities.
• Tool Calling Support – Detects models supporting function calling and tool use.
• JSON Output Support – Identifies models with JSON output capabilities.
• Context Length Tracking – Includes maximum context window information.
• Token Limits – Captures maximum prompt, completion, and response token limits.
• Architecture Information – Extracts model architecture, tokenizer, and instruction type.
• Moderation Status – Indicates if Open Router Model includes safety moderation.
• Multi-Provider Support – Covers models from OpenAI, Anthropic, Cohere, Meta, and more.
• Date Tracking – Records model creation date for version tracking.
• Direct Links – Generates direct OpenRouter links to each Open Router Model.
• Proxy Support – Apify residential proxy support for reliable API access.
• Bulk Model Processing – Discovers and analyzes multiple Open Router Model simultaneously.
• Real-Time Dataset Push – Pushes results to Apify Dataset with metadata.
• Timestamp Recording – Records scrape timestamp for audit trails.

---

🌍 Data Source

OpenRouter API

• Authority: OpenRouter.ai - AI model aggregation platform
• Coverage: 200+ AI models from multiple providers
• Data Quality: Official, real-time pricing and model data
• API: REST API v1
• Endpoint: https://openrouter.ai/api/v1/models
• Authentication: No authentication required for public data
• Response Format: JSON
• Rate Limits: Reasonable rate limits for research use
• Completeness: Includes LLMs, vision models, and specialized models

---

⚙️ How It Works

The Open Router Model Scraper connects to the OpenRouter API and retrieves the complete catalog of available Open Router Model options. It filters results based on optional keywords, extracts comprehensive model metadata, calculates pricing information, and detects model capabilities. Results are pushed to the Apify Dataset with complete information.

Key Processing Steps:

1. Input Parsing – Accept keywords and configuration from Actor input
2. Proxy Setup – Configure Apify residential proxy if available
3. API Connection – Connect to OpenRouter API with proper headers
4. Model Fetch – Retrieve complete catalog of Open Router Model options
5. Keyword Filtering – Filter models based on provided keywords
6. Model Parsing – Extract comprehensive metadata from each model
7. Pricing Calculation – Convert pricing from API format to readable rates
8. Capability Detection – Identify vision, tools, and JSON capabilities
9. Date Formatting – Parse and format creation dates
10. Result Compilation – Aggregate all model information
11. Dataset Push – Push individual model records to Apify Dataset
12. Progress Logging – Report processing statistics

Key Benefits:

• Discover all available Open Router Model options
• Compare Open Router Model pricing across providers
• Identify models with specific capabilities
• Research Open Router Model architecture and specifications
• Find cost-effective Open Router Model alternatives
• Plan AI application development with accurate pricing
• Track Open Router Model updates and releases

---

📥 Input

The Actor accepts the following input parameters:

Field	Type	Default	Description
keywords	string	"all"	Comma-separated keywords to filter Open Router Model (max 10). Use "all" for complete catalog. Examples: "gpt,claude,vision"
useApifyProxy	boolean	true	Enable Apify residential proxies for API access
apifyProxyGroups	array	["RESIDENTIAL"]	Proxy group configuration

Example Input:

{
  "keywords": "claude,vision",
  "useApifyProxy": true,
  "apifyProxyGroups": ["RESIDENTIAL"]
}

All Models Example:

{
  "keywords": "all",
  "useApifyProxy": true
}

Multi-Keyword Example:

{
  "keywords": "gpt-4,claude-3,vision,multimodal",
  "useApifyProxy": true
}

---

📤 Output

The Actor pushes Open Router Model records with the following structure:

Field	Type	Description
keyword	string	Search keyword used to find this model
model_id	string	Full model identifier (provider/model-name)
provider	string	Model provider (OpenAI, Anthropic, etc.)
model_name	string	Model name without provider prefix
display_name	string	Human-readable model display name
description	string	Model description (300 chars max)
created_date	string	Model creation/release date (YYYY-MM-DD)
context_length	integer	Maximum context window size
max_completion_tokens	integer	Max completion tokens
max_prompt_tokens	integer	Max prompt tokens
max_response_tokens	integer	Max response tokens
price_per_1m_input	string	Input token pricing per million tokens
price_per_1m_output	string	Output token pricing per million tokens
price_per_image	string	Image processing pricing
price_per_request	string	Per-request pricing (if applicable)
modality	string	Model type (text, text+vision, etc.)
tokenizer	string	Tokenizer type used
instruct_type	string	Instruction following format
supports_vision	string	Vision/image capability (yes/no)
supports_tools	string	Function calling/tool support (yes/no)
supports_json_output	string	JSON output capability (yes/no)
is_moderated	string	Safety moderation status
openrouter_url	string	Direct link to model on OpenRouter
scraped_at	string	ISO 8601 scrape timestamp

Example Output Record (Claude):

{
  "keyword": "claude",
  "model_id": "anthropic/claude-3-opus",
  "provider": "anthropic",
  "model_name": "claude-3-opus",
  "display_name": "Claude 3 Opus",
  "description": "Most powerful Claude model, best for complex tasks requiring nuance, creativity, and reasoning.",
  "created_date": "2024-02-29",
  "context_length": 200000,
  "max_completion_tokens": 4096,
  "max_prompt_tokens": 200000,
  "max_response_tokens": 4096,
  "price_per_1m_input": "$0.0150",
  "price_per_1m_output": "$0.0750",
  "price_per_image": "",
  "price_per_request": "",
  "modality": "text",
  "tokenizer": "claude",
  "instruct_type": "chat",
  "supports_vision": "yes",
  "supports_tools": "yes",
  "supports_json_output": "yes",
  "is_moderated": "True",
  "openrouter_url": "https://openrouter.ai/anthropic/claude-3-opus",
  "scraped_at": "2025-02-14T12:00:00Z"
}

Example Output Record (Vision Model):

{
  "keyword": "vision",
  "model_id": "openai/gpt-4-vision-preview",
  "provider": "openai",
  "model_name": "gpt-4-vision-preview",
  "display_name": "GPT-4 Vision",
  "description": "Multimodal model supporting image inputs alongside text for comprehensive understanding.",
  "created_date": "2024-01-15",
  "context_length": 128000,
  "max_completion_tokens": 4096,
  "max_prompt_tokens": 128000,
  "max_response_tokens": 4096,
  "price_per_1m_input": "$0.0100",
  "price_per_1m_output": "$0.0300",
  "price_per_image": "$0.0086",
  "price_per_request": "",
  "modality": "text+vision",
  "tokenizer": "cl100k_base",
  "instruct_type": "chat",
  "supports_vision": "yes",
  "supports_tools": "yes",
  "supports_json_output": "yes",
  "is_moderated": "True",
  "openrouter_url": "https://openrouter.ai/openai/gpt-4-vision-preview",
  "scraped_at": "2025-02-14T12:00:00Z"
}

---

🧰 Technical Stack

• HTTP Requests: requests library with session management
• API: OpenRouter REST API v1 (JSON)
• Async: asyncio for concurrent operations
• JSON: Native JSON parsing
• Proxy: Apify Proxy with residential support
• Logging: Apify Actor logging system
• Platform: Apify Actor serverless environment
• Timeout: 15 seconds per API request

---

📊 Data Fields Explained

Model Identification

• model_id: Unique identifier (provider/model-name)
• provider: Model provider (OpenAI, Anthropic, etc.)
• model_name: Model name without provider
• display_name: Human-readable name

Capabilities

• context_length: Maximum context window
• supports_vision: Image/vision processing capability
• supports_tools: Function calling support
• supports_json_output: JSON-mode output capability
• modality: Model type (text, text+vision, audio, etc.)

Token Limits

• max_completion_tokens: Maximum response tokens
• max_prompt_tokens: Maximum input tokens
• max_response_tokens: Maximum response length

Pricing

• price_per_1m_input: Per-million-token input cost
• price_per_1m_output: Per-million-token output cost
• price_per_image: Image processing cost
• price_per_request: Fixed per-request cost

Technical Details

• tokenizer: Tokenization method
• instruct_type: Instruction format (chat, completion, etc.)
• modality: Model type
• is_moderated: Safety moderation applied

Metadata

• created_date: Model release date
• description: Model overview
• openrouter_url: Direct OpenRouter link
• scraped_at: Data retrieval timestamp

---

🎯 Model Capabilities

Vision Support

Models with image/vision input capabilities:

• GPT-4 Vision
• Claude 3 (all variants)
• LLaVA models
• Llama 2 Vision

Tool Calling

Models supporting function calling:

• GPT-4 (all variants)
• Claude 3 (all variants)
• Llama 2
• Mistral

JSON Output

Models with JSON-mode:

• GPT-4 (all)
• Claude 3 (all)
• Llama 2
• Mistral

---

🎯 Use Cases

• AI Model Research – Research available Open Router Model options
• Cost Comparison – Compare Open Router Model pricing across providers
• Application Selection – Find optimal Open Router Model for specific needs
• Capability Discovery – Identify models with required features
• Pricing Analysis – Analyze Open Router Model cost structures
• Provider Evaluation – Compare different providers' Open Router Model offerings
• Performance Planning – Select models based on context length and token limits
• Feature Requirements – Find models supporting vision, tools, or JSON
• Budget Planning – Estimate costs using Open Router Model pricing data
• Competitive Analysis – Track competing AI models on OpenRouter
• AI Stack Selection – Choose models for multi-model AI applications
• Testing Alternatives – Find alternative Open Router Model for testing
• Vendor Comparison – Compare OpenAI, Anthropic, Cohere, etc.
• Model Benchmarking – Compare model specifications and capabilities
• Pricing Alerts – Monitor Open Router Model pricing changes

---

🚀 Quick Start

1. Prepare Input

Go to Apify Console and enter:

{
  "keywords": "claude,gpt-4,vision",
  "useApifyProxy": true,
  "apifyProxyGroups": ["RESIDENTIAL"]
}

2. Run the Actor

Click Start button. The Actor will:

• Connect to OpenRouter API
• Filter models by keywords
• Extract comprehensive model data
• Parse pricing information
• Detect capabilities
• Push results to Dataset

3. Monitor Progress

Console shows:

Keywords (3): claude,gpt-4,vision
Proxy active: ['RESIDENTIAL']
Total models on OpenRouter: 287
Keyword 'claude' → 12 models matched
  Pushed 12 records for 'claude'
Keyword 'gpt-4' → 8 models matched
  Pushed 8 records for 'gpt-4'
Keyword 'vision' → 15 models matched
  Pushed 15 records for 'vision'
Done! 3 keyword(s), 35 total records pushed.

4. View & Download Results

• Results Tab: All model records
• Export: JSON, CSV, Excel
• Filter: By provider or capability
• Sort: By pricing or context length

---

📈 Performance

Processing Speed

• ~2-5 seconds for single API fetch
• ~10-30 seconds for parsing and pushing 200+ models
• Includes keyword filtering and capability detection

Resource Usage

• Memory: ~50-80MB
• CPU: ~10-15% during processing
• Network: ~500KB-2MB per full scan
• API calls: 1 (single request to fetch all models)

Scalability

• Handles 200+ models efficiently
• Real-time Dataset push
• No pagination needed (API provides all models in single response)

---

⚠️ Important Notes

Legal & Compliance

• Fair Use: Data collection respects OpenRouter terms
• Public Data: All data from public OpenRouter API
• Attribution: OpenRouter provides all model data
• Pricing: Subject to change; verify with OpenRouter for current rates
• Terms of Service: Complies with OpenRouter ToS

Data Quality

• Accuracy: Reflects OpenRouter's live catalog
• Currency: Pricing updated by OpenRouter
• Completeness: Includes all models in OpenRouter catalog
• Freshness: Real-time data from API
• Verification: Always verify pricing on OpenRouter

Best Practices

• Verify pricing before integration
• Check context length requirements
• Test model capabilities for your use case
• Monitor for model updates
• Compare with latest OpenRouter data
• Consider model performance/cost tradeoffs

---

---

📦 Changelog

Initial Release:

• OpenRouter API v1 integration
• Complete Open Router Model catalog fetching
• Multi-keyword filtering (up to 10 keywords)
• Comprehensive metadata extraction (20+ fields)
• Pricing calculation and formatting
• Vision support detection
• Tool calling capability detection
• JSON output support identification
• Context length tracking
• Token limit extraction
• Model architecture information
• Moderation status detection
• Date parsing and formatting
• Direct URL generation
• Apify proxy support
• Bulk model processing
• Real-time Dataset push
• ISO 8601 timestamp recording
• Error handling and logging
• Asyncio executor for non-blocking requests

---

🧑‍💻 Support & Feedback

• Issues: Submit via Apify console
• Documentation: Check Actor details page
• Community: Apify forum discussions
• Feature Requests: Suggest improvements
• Bug Reports: Include keywords and errors

Output Access

• Results Tab: All model records
• Export: JSON, CSV, Excel
• Filter: By provider or capability
• API: Query via Apify API

---

📄 License & Legal

Terms of Use:

• Use for legitimate AI research and development
• Respect OpenRouter terms of service
• Verify pricing before integration
• Don't republish without attribution
• Comply with applicable laws
• Use data ethically and responsibly

Disclaimer: Open Router Model Scraper is provided as-is for research purposes. Users are responsible for ensuring compliance with OpenRouter ToS. Always verify model data and pricing with official OpenRouter sources.

---

🎉 Get Started Today

Deploy now for AI model research!

Use for:

• 🤖 Model Research
• 💰 Pricing Analysis
• 🔍 Capability Discovery
• 📊 Provider Comparison
• 🚀 Application Selection

Perfect for:

• AI Developers
• ML Engineers
• Product Managers
• AI Researchers
• Startup Founders

---

Last Updated: February 2025
Version: 1.0.0
Status: Production Ready
Platform: Apify Actor
Architecture: Async/Await
API Source: OpenRouter v1
Models Tracked: 200+
Data Quality: Real-time

---

📚 Related Tools

• Smart Article Extractor
• Skill Curator Scraper
• Website Technology Stack Scraper
• Google Keyword Finder

Your complete Apify-powered AI model discovery solution! 🚀✨

---

🤖 AI Model Excellence

This Actor is optimized for Open Router Model discovery with:

• ✅ Complete OpenRouter API integration
• ✅ Multi-keyword filtering
• ✅ Comprehensive field extraction
• ✅ Pricing calculation
• ✅ Capability detection
• ✅ Real-time Dataset integration
• ✅ Error recovery
• ✅ Production-ready code

Discover AI models effortlessly! 💎🚀