LLM Benchmark Leaderboard Scraper

Scrape AI model benchmark data from leading LLM evaluation platforms. Get intelligence scores, coding performance, speed metrics, pricing, and provider information for every major AI model.

Features

• Multiple sources: Artificial Analysis (comprehensive benchmarks) and Open LLM Leaderboard (HuggingFace)
• Intelligence Index: Overall model performance scores
• Coding Index: Code generation capability scores
• Speed metrics: Output tokens per second
• Pricing data: Input and output token costs per model
• Model metadata: Parameters, context length, license, open-source status

Input

Field	Type	Default	Description
source	select	artificialanalysis	Data source: artificialanalysis, open-llm-leaderboard, or all
limit	integer	50	Max number of models to return (1-500)

Sample Input

{
  "source": "artificialanalysis",
  "limit": 100
}

Output

{
  "source": "artificialanalysis",
  "model_name": "GPT-5.6 Sol",
  "organization": "OpenAI",
  "intelligence_index": 85.2,
  "coding_index": 82.1,
  "speed": 125.5,
  "input_price": 15.0,
  "output_price": 60.0,
  "open_source": false,
  "context_length": 128000,
  "parameters": "",
  "license": "",
  "scraped_at": "2026-07-14T20:00:00.000Z"
}

Use Cases

• AI model selection: Compare models before choosing one for your project
• Market intelligence: Track which AI labs are winning the benchmark race
• Content creation: Generate "top AI models" articles and videos with real data
• Investment research: Monitor AI landscape trends for investment signals
• Cost optimization: Find the best price/performance ratio for your use case
• Research: Track benchmark progress over time

Pricing

Pay-per-result: $1.00 per 1,000 results

Links

• Artificial Analysis
• Open LLM Leaderboard