Universal Intelligence Benchmark

Inference-agnostic intelligence measurement. You provide the API key. We measure the intelligence.

Your API key stays in your browser. We orchestrate the evaluation pipeline — not the inference.

Configuration

API Provider

API Base URL

API Key

Model

Use OpenRouter model IDs like openai/gpt-4o or anthropic/claude-3.5-sonnet

Dimensions

Mode: Quick (40 tasks, ~2 min) Full (200 tasks, ~10 min)

Results

—

Composite Score

—

Efficiency (score/cost)

Dimension	Score	Tasks	Latency	Cost

Total cost: — Total time: —

Leaderboard

Loading leaderboard…

Connected Research

AI EconomicsBenchmark scores vs real ROI Cost-Effective AICheapest model often wins HPF-P FrameworkDecision readiness as intelligence proxy Open HumanoidEmbodied intelligence validation Future of AIWhen benchmarks become obsolete

API Documentation

# Check service status
curl -H "X-API-Key: YOUR_STABILARITY_KEY" \
  https://hub.stabilarity.com/api/v1/uib/status

# Run benchmark
curl -X POST \
  -H "X-API-Key: YOUR_STABILARITY_KEY" \
  -H "Content-Type: application/json" \
  -d '{"api_key":"YOUR_OPENROUTER_KEY","model":"openai/gpt-4o","quick_mode":true}' \
  https://hub.stabilarity.com/api/v1/uib/run

# Get leaderboard
curl -H "X-API-Key: YOUR_STABILARITY_KEY" \
  https://hub.stabilarity.com/api/v1/uib/leaderboard