Perplexity Decider v1.2 Release - DI 0.3 Text - Joint Requests: leaderboard

Decision Index 0.3 (public); overall "uses the suite's weighting, not a simple category average"; own runs (card). Card: soup "chosen as the best compromise across all our evaluations"; training includes official Decision Index training splits "decontaminated against the test sets". not stated (public suite) Follow existing release naming ("WeMM-JEV Release - DI 0.3 Text"). Do not merge with jevalkit DecisionIndex_Text (0.4.1 public, board import) or any Decision Index 0.4.1 Full score. Metric: overall public Decision Index 0.3 (suite weighting).

Metric: overall public Decision Index 0.3 (suite weighting). Source: huggingface.co. Saturation forecast: Rough model projection: around 2026. 2 models tracked. Updated 14 Nov 2026.

Top models

#ModelScoreOverall rank
1PPLX-Decider-v1.268.09
2PPLX-Decider-Flash-v1.258.88

Interactive version: theaggregate.ai/benchmark?slug=perplexity-decider-v1-2-release-di-0-3-text-joint-requests · How It Works · Data refreshed daily, snapshot 2026-11-16.