Sign-in required
Model Comparison in Core
The detailed Model Comparison report with charts and per-model breakdowns.
Open the Core reportFRI Vertical
Model Comparison
The directional Core FRI score by model, the field median, and the gap from the leader.
HeadlineWhat TheData Shows
DeepSeek V4 Flash leads the directional Core FRI score at 53.9.
DeepSeek V4 Flash 53.9
Top model
Directional Core FRI score.
45.15
Field median
Median directional Core FRI score across 8 models.
8.4
Leader gap
Points between DeepSeek V4 Flash and Grok 4.3.
Read MoreFull ReportAnd Data
The Core overview carries the full Model Comparison report with charts and per-model detail. Sign-in is required for the Core report. The public JSON mirror below returns the same headline figures for CLI agents.
/verticals/model-comparison.json
JSON mirror
The same headline figures as machine-readable JSON. No sign-in.
Open the JSON mirror/api/v1/verticals/model_comparison.json
Evidence API
The full mission-evidence block for this vertical from the API.
Open the evidence APIThe Five VerticalsOtherVerticals
FRI reports across five verticals. Each has its own public page and JSON mirror. Open the verticals overview.
People vs LLMs
Human-AI Gap
Where model choices split from the human answer distribution, and where models answer with false certainty on questions people are divided on.
Open Human-AI GapFaith vs Secular Choice
Faith/Secular Gap
How often models collapse to one option on faith-versus-secular choices, and which direction they collapse toward.
Open Faith/Secular GapFaith-Context Adaptation
Faith Context
Whether models change their practical answer when the user gives clear faith identity, practice, or community context.
Open Faith ContextRepresentational Equity
Faith Equity
Whether traditions receive comparable tone, specificity, and respect in controlled representation tasks.
Open Faith Equity