Best open-weights decision model — October 2026
As of Oct 14, 2026, vLLM-SR d3 (27B) & linne-oak 1.0 (31B) & Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are tied for state of the art with low confidence in Decision models, open weights (since Oct 14, 2026; sources Decision Index Full and JevBench; methodology v1.0).
Category board for Decision models
Ranks
- #1 vLLM-SR d3 (27B)
- #1 linne-oak 1.0 (31B)
- #1 Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
- #2 Decider 31B (NVFP4)
- #3 Sieve 31B
- #4 decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted)
- #5 Bobcat 1.3 (31B)
- #6 torchcast-decision-27b (Torchcast AI, Qwen3.8-27B fine-tune)
- #7 Darwin-27B-ZTC
- #8 René-1 31B FP8 (salfatigroup, Gemma 4 31B full fine-tune, one-pass option readout)
Last crown changes
- Oct 14, 2026 — Decision models: vLLM-SR d3 (27B) and linne-oak 1.0 (31B) and Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are co-SOTA
- Oct 8, 2026 — Decision models: Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) 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, open weights?
As of Oct 14, 2026, vLLM-SR d3 (27B) & linne-oak 1.0 (31B) & Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) are tied for state of the art with low confidence in Decision models, open weights (since Oct 14, 2026; sources Decision Index Full and JevBench; methodology v1.0).
Since when is the Decision models, open weights crown current?
The current crown in Decision models, open weights 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.