Who is SOTA in Decision models?Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) & REAX 1.3 & GLiDE 27B
Typed answers with probabilities. The crown follows Decision Index Full and JevBench, both independent at weight 1. Vendor-reported numbers are display-only at weight 0. As of Nov 6, 2026, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) & REAX 1.3 &… Crown history, benchmark weights and methodology.
Badges for the leaders
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
[](https://isitsota.com/m/pplx-decider-v1-1-27b)
<a href="https://isitsota.com/m/pplx-decider-v1-1-27b"><img alt="CO-SOTA · Decision models · Nov 2026" src="https://isitsota.com/badge/pplx-decider-v1-1-27b/decision.svg" height="28"></a>
[](https://isitsota.com/m/reax-1-3)
<a href="https://isitsota.com/m/reax-1-3"><img alt="CO-SOTA · Decision models · Nov 2026" src="https://isitsota.com/badge/reax-1-3/decision.svg" height="28"></a>
[](https://isitsota.com/m/glide-v1-2-1-loop)
<a href="https://isitsota.com/m/glide-v1-2-1-loop"><img alt="CO-SOTA · Decision models · Nov 2026" src="https://isitsota.com/badge/glide-v1-2-1-loop/decision.svg" height="28"></a>
Reliability · Turion Jev 31B
Confidence on the horizontal axis, accuracy on the vertical axis. Larger marks carry more weight.
Self-reported claims
No vendor-reported row is in this snapshot. Vendor-reported numbers are display-only at weight 0.
Best page and recent changes
Best AI model for decision models
- Decision models: REAX 1.3 and GLiDE 27B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA
- Decision models: REAX 1.3 and GLiDE 27B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA
- Decision models: GLiDE 27B and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA
What can move this crown
- Decision Index Full — primary, weight 1
- JevBench — primary, weight 1
- Decision Index latency — display, weight 0 (display only, cannot crown)
- JevBench price — display, weight 0 (display only, cannot crown)
- Vendor decision claims — display, weight 0 (display only, cannot crown)
Who is SOTA in Decision models?
As of Nov 6, 2026, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) & REAX 1.3 & GLiDE 27B are tied for state of the art with low confidence in Decision models (since Nov 6, 2026; sources JevBench and Decision Index Full; methodology v1.0).
Since when is the Decision models crown current?
The current crown in Decision models has held since Nov 6, 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.