Skip to content

BenchGecko Documentation

Every model. Every agent. Everything AI. Tracked.

BenchGecko is the data layer of the AI economy. Thousands of models with cross-provider pricing and daily price history. Hundreds of companies with valuations, funding timelines, and revenue estimates. Benchmark scores, developer adoption signals, agent leaderboards, and a changelog that captures every price drop, every launch, every deprecation as it happens.

The Platform at a Glance

BenchGecko organizes the AI landscape into ten interconnected data layers:

Layer What It Tracks Explore
Models Thousands of AI models with benchmark scores across 128 evaluations benchgecko.ai/models
Providers Hundreds of inference providers with availability and uptime benchgecko.ai/providers
Pricing Cross-provider pricing with arbitrage finder and price history benchgecko.ai/pricing
Compare Side-by-side model comparison across all dimensions benchgecko.ai/compare
Benchmarks 128 evaluation suites with score evolution over time benchgecko.ai/benchmarks
Agents 165 AI agents scored on SWE-bench, GAIA, OSWorld, and more benchgecko.ai/agents
MCP Servers 4,923 Model Context Protocol servers rated and categorized benchgecko.ai/mcp
Economy AI Bubble Index, valuations, funding, market cap for 71 companies benchgecko.ai/economy
Compute Supply chain across foundries, chips, memory, systems, and energy benchgecko.ai/compute
Mindshare Developer attention across Reddit, HN, GitHub, arXiv, and X benchgecko.ai/mindshare

Key Indicators

BenchGecko produces several proprietary composite indices:

  • Gecko Pulse: A 0-100 health score of the AI ecosystem. Blends the inverse Bubble Index, benchmark velocity, price compression, mindshare diversity, and supply-chain tension.
  • AI Bubble Index: Measures whether AI valuations are justified by fundamentals. Currently reading 278% (frothy).
  • AICDI: AI Compute Demand Index tracking infrastructure pressure across five supply chain layers.
  • Mindshare Pulse: Narrative energy distribution across models, agents, companies, and topics.

Getting Started

For Developers

The BenchGecko API provides free programmatic access to model data, benchmark scores, and pricing. No API key required for the free tier.

import requests

response = requests.get("https://benchgecko.ai/api/v1/models?sort=score&limit=5")
models = response.json()

for model in models["data"]:
    print(f"{model['name']}: {model['average_score']} avg, ${model['input_price']}/M in")

Full API documentation: benchgecko.ai/api-docs

For Researchers

BenchGecko maintains open datasets on Harvard Dataverse, Figshare, and Hugging Face. Every page includes a Cite & Share bar with APA, MLA, BibTeX, Chicago, and plain-text citation formats.

For Business

Use the Pricing Calculator to estimate monthly inference costs, the Compare tool to evaluate models against your requirements, and the Economy Dashboard to track the competitive landscape.

Data Sources

BenchGecko aggregates data from 10+ authoritative sources:

All data is refreshed every 2 hours from primary sources. Zero editorial content.