SO-Arena v1.0 - Null Discrimination: leaderboard
SO-Arena v1.0 native null discrimination. Stability measures consistency under option reordering, null discrimination measures missing-answer handling, and reliability combines the two. The published JSON supplies these exact values; StartLux null scores are derived from rounded reliability values. Native model inference paths and API interfaces remain separate. Metric: Native null discrimination (0-1). Higher is better.
Metric: Native null discrimination (-1 to 1). Source: raw.githubusercontent.com. Saturation forecast: Rough model projection: around 2026. 19 models tracked. Updated 14 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | JevAny-27B | 0.7934 | ||
| 2 | Jev 1.14.0 | 0.7794 | ||
| 3 | Kev 27B | 0.764 | ||
| 4 | StartLux-v1.1-4B-AutoThink | 0.7555 | ||
| 5 | KnowLine-31B-Gen1 | 0.7104 | ||
| 6 | Sieve-31B-v3 | 0.7069 | ||
| 7 | StartLux-v1.1-4B-Instant | 0.6991 | ||
| 8 | GLiDE v1.2 | 0.6384 | ||
| 9 | Clef 27B | 0.6319 | ||
| 10 | PPLX-Decider-v1.2 | 0.6134 | ||
| 11 | JevAny-4B | 0.6024 | ||
| 12 | Clef flash 9.5B | 0.5913 | ||
| 13 | GPT-6 Luna | SO-Arena run: Decisions API; model=gpt-6-luna | 0.55 | #99 |
| 14 | PPLX-Decider-Flash-v1.2 | 0.508 | ||
| 15 | Kev 4B | 0.5036 |
Interactive version: theaggregate.ai/benchmark?slug=so-arena-v1-0-null-discrimination · How It Works · Data refreshed daily, snapshot 2026-11-16.