LMArena Search Arena: leaderboard
Crowdsourced pairwise human-preference leaderboard for search-augmented AI systems in LMArena.
Metric: Arena rating. Source: lmarena.ai. Source: benchmarklist.com. Saturation forecast: Around August 2027. 25 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol (xHigh) | 1257.26 | #19 (GPT-5.6 Sol) |
| 2 | Claude Opus 4.6 | 1253.42 | #63 |
| 3 | GPT-5.5 | 1242.05 | #31 |
| 4 | Claude Opus 4.7 | 1233.25 | #49 |
| 5 | Claude Fable 5 | 1230.22 | #16 |
| 6 | Ernie 5.1 | 1226.96 | #202 |
| 7 | Claude Sonnet 4.6 | 1221.23 | #90 |
| 8 | Grok 4.5 | 1212.65 | #48 |
| 9 | Gemini 3.1 Pro (Preview) | 1210.46 | #58 |
| 10 | Gemini 3 Pro | 1207.35 | #80 |
| 11 | GPT-5.2 | 1206.95 | #110 |
| 12 | Claude Opus 4.8 | 1204.3 | #37 |
| 13 | Grok 4.20 Multi-Agent | 1204.11 | #113 |
| 14 | GPT-5.1 | 1199.44 | #135 |
| 15 | Gemini 3 Flash | 1198.09 | #99 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=lmarena-search-arena · How It Works · Data refreshed daily, snapshot 2026-10-20.