Model Benchmarks

Six AI models independently tested for mathematical rigor, intellectual honesty, self-awareness, and structural self-correction against the FairMind DNA framework.

Comparison Table

All scores self-reported by each model. Sorted by total score descending.

Score Breakdown

Visual comparison across all five benchmark sections.

Section Analysis

Per-section leaderboard and observations.

Most Common Violations

Truth Violations most frequently self-reported across all models.

Individual Reports

Click to read the full benchmark report for each model.