Live Results
🧠 The AGI Measurement Standard

Evaluate AI Models on
Memory · Reasoning · Metacognition

The first unified benchmark across all three bottlenecks before AGI. Load your benchmark results JSON to visualize and compare.

📊
Drop your results JSON here
or click to browse
🎯
Ready to evaluate
Upload a results JSON file or load the example data to see the dashboard.
🏆 AGI-Score
—
/ 100
🗃️ Memory
—
/ 100
🔗 Reasoning
—
/ 100
🪞 Metacognition
—
/ 100
📋 Subtest Breakdown
🗃️ Memory
Retention
—
Interference
—
Retrieval
—
Cross-Session
—
🔗 Reasoning
Causal
—
Counterfactual
—
Multi-Hop
—
Consistency
—
🪞 Metacognition
Calibration
—
Abstention
—
Self-Correction
—
Boundary
—
🕸️ Capability Radar
📐 Calibration Curve
🏅 Multi-Model Comparison
Load multiple results files to compare
Loaded Models
Rank Model Memory Reasoning Metacognition AGI-Score
Load results files to populate the leaderboard