SO-Arena v1.0 - Overall: leaderboard
SO-Arena v1.0 overall equal-weight mean of11domain scores on the source-selected1232-task panel. Domain scores average native task credits for static decisions, tool outcomes or web progress, then scale to0–100. The full release has2919tasks; this leaderboard emphasizes the highest difficulty tiers and keeps all untiered sentiment/knowledge tasks. Metric: Equal-domain macro credit (%). Higher is better.
Metric: Equal-domain macro credit (%). Source: huggingface.co. Saturation forecast: Around February 2027. 20 models tracked. Updated 14 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-6 Sol | Medium | 82.32 | #26 |
| 2 | GPT-6 Luna | Medium | 79.68 | #99 |
| 3 | GLiDE v1.2 | 70.21 | ||
| 4 | KnowLine-31B-Gen1 | 70.19 | ||
| 5 | Sieve-31B-v3 | 67.65 | ||
| 6 | PPLX-Decider-v1.2 | 66.7 | ||
| 7 | Kev 27B | 61.39 | ||
| 8 | Jev 1.14.0 | 60.43 | ||
| 9 | StartLux-v1.1-4B-Instant | 58.96 | ||
| 10 | StartLux-v1.1-4B-AutoThink | 58.63 | ||
| 11 | JevAny-27B | 58.37 | ||
| 12 | Clef 27B | 56.87 | ||
| 13 | H2O-Lightning-4B | 54.86 | ||
| 14 | PPLX-Decider-Flash-v1.2 | 52 | ||
| 15 | Clef flash 9.5B | 51.61 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=so-arena-v1-0-overall · How It Works · Data refreshed daily, snapshot 2026-11-16.