Sign in
VERIFIEDNov 222026

Is karpo-decision-31B state of the art?Yes.SOTA in Decision models (tied).

As of Nov 22, 2026, karpo-decision-31B is tied for state of the art with low confidence in Decision models (since Nov 11, 2026; sources Decision Index Full; methodology v1.0). karpo-decision-31B is co-SOTA in Decision models with low confidence because primary sources disagree beyond the margin of error (with GLiDE 27B, kudhab (Gemma 4 31B), Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)).

Computed automaticallyas of Nov 22, 2026methodology v1.0confidence: low
View as of point-in-time crowns
At a glance

Facts and scores

karpo-decision-31B is an open-weights text model from MachinePulse; 31B parameters and no public price.

How it scores

  • On Decision Index Full it scores 73.9 (CI 73.4–74.4), #1 of 420 (top 1%); GLiDE 27B is 0.5 below.
  • On Decision Index Brier it scores 23.9%, #5 of 419 (top 2%); 3.5 pts behind lg-decision-4B; level with StartLux-Decision v1.1 27B above and Strands Decider v2 31B is 0.2 pts below.
  • On Decision Index ECE it scores 4.8%, #135 of 420 (top 33%); 3.7 pts behind KnowLine 9B Gen2; level with AJev lora1 (Gemma 4 26B-A4B) above and level with munku (Gemma 4 31B) below.

History

First result on isitsota: Nov 21, 2026. Best overall rank so far: #1 in Decision models, first held Nov 11, 2026. Crown intervals: Decision models Nov 11, 2026–Nov 12, 2026, Decision models Nov 12, 2026–Nov 21, 2026 and Decision models Nov 21, 2026–now.

Calibration

Reliability

Index bins for karpo-decision-31B. Confidence on the horizontal axis, accuracy on the vertical axis.

Median 94.3 ms, same GPU.

Receipts

Evidence

Scores come from the latest ingested snapshot
SourceScoreDateTypeReceipt
Decision Index Full
karpo-decision-31B
73.9 73.4–74.42026-11-21INDEP.snap_decision-index_2026-11-21_release-v4|decision-index-0.4.1
Reign history

Where it sits in the succession

Full timeline →
Decision models
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
MISA AI 2.0 Decision Plus · oak (31B)
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
vLLM-SR d3 (27B)
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
StartLux-Decision v1.1 27B
baselabs mammoth-1 (27B)
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
Fastino GLiDE 1.2
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
Fastino GLiDE 1.2
KnowLine 31B Gen1
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
Sieve 31B v3
GLiDE 27B
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
REAX 1.3
GLiDE 27B
karpo-decision-31B
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
REAX 1.3
GLiDE 27B
karpo-decision-31B
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
GLiDE 27B
karpo-decision-31B
kudhab (Gemma 4 31B)
Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)
Links

Related pages

Is karpo-decision-31B SOTA?

As of Nov 22, 2026, karpo-decision-31B is tied for state of the art with low confidence in Decision models (since Nov 11, 2026; sources Decision Index Full; methodology v1.0).

What is karpo-decision-31B best at?

As of Nov 22, 2026, karpo-decision-31B is tied for state of the art with low confidence in Decision models (since Nov 11, 2026; sources Decision Index Full).

How does karpo-decision-31B rank in Decision models?

Best current rank: #1 in Decision models, as of Nov 22, 2026.

When was karpo-decision-31B last SOTA?

As of Nov 22, 2026, karpo-decision-31B last held Decision models from Nov 11, 2026 until Nov 12, 2026.