Most calibrated decision model — October 2026
As of Oct 15, 2026, Nace.AI Drex v2.0 (35B-A3B) is state of the art in Decision models, calibration (since Oct 14, 2026; sources Decision Index ECE; methodology v1.0).
Category board for Decision models
Ranks
Last crown changes
- Oct 14, 2026 — Decision models: Nace.AI Drex v2.0 (35B-A3B) takes the crown
Which benchmarks count
- Decision Index Full — primary, weight 1
- JevBench — primary, weight 1
- Decision Index latency — display, weight 0
- JevBench price — display, weight 0
- Vendor decision claims — display, weight 0
Who is SOTA in Decision models, calibration?
As of Oct 15, 2026, Nace.AI Drex v2.0 (35B-A3B) is state of the art in Decision models, calibration (since Oct 14, 2026; sources Decision Index ECE; methodology v1.0).
Since when is the Decision models, calibration crown current?
The current crown in Decision models, calibration has held since Oct 14, 2026.
What is a decision model?
A decision model takes state, as text or as images, plus a typed question. It returns a typed answer with probabilities rather than free text. Output types: noul (yes/no/unknown), choice, score, multi-label.
How is calibration ranked?
The calibration slice ranks the lowest ECE among models in the top quartile of accuracy, so a constant probability cannot take the slice. Brier breaks ECE ties.
Do vendor-reported decision numbers count?
Vendor-reported numbers are display-only at weight 0.