LMArena Search leaderboard
As of 6 Nov 2026, the top score on LMArena Search is 1258 by GPT-5.6 Sol (OpenAI), ahead of GPT-6.1 Sol at 1250, across 41 model configurations with a published result (Arena (LMArena), result dated 13 Oct 2026). Search Arena compares models that answer with live web search, judged on the answer and its sourcing. LMArena Search is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- GPT-5.6 Sol holds the top Arena Search score at 1258 (Arena (LMArena), 13 Oct 2026).
- GPT-6.1 Sol is second on Arena Search at 1250 (Arena (LMArena), 13 Oct 2026).
- Arena Search: Pairwise votes on answers that used web search (Arena (LMArena)).
- 41 model configurations have a published Arena Search result on BenchLeader (BenchLeader, 6 Nov 2026).
- Arena Search does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Pairwise votes on answers that used web search.
- Published by
- Arena (LMArena)
- Category
- Agents & tools
- Index weight
- Reference only
- Models
- 41
- Data as of
- 6 Nov 2026
CC BY 4.0 — lmarena-ai/leaderboard-dataset on Hugging Face.
What the test looks like
Search Arena compares models that answer with live web search, judged on the answer and its sourcing.
How it is scored
Bradley-Terry rating from pairwise votes, in Elo-like units, with style control.
What to keep in mind
Only search-enabled products are listed, and results depend on the day.
- 1GPT-5.6 Sol (xhigh)1258
- 2GPT-6.1 Sol (max)1250
- 3GPT-6 Astra (max)1246
- 4Claude Fable 5.1 (high)1233
- 5Qwen3.8 Max (max)1230
- 6Claude Fable 5 (high)1229
- 7GPT-6 Sol (max)1228
- 8GPT 5.5 Search1227
- 9Claude Opus 4.6 Search1222
- 10Claude Opus 5 (high)1214
- 11Claude Opus 5.5 (medium)1211
- 12Claude Opus 4.71210
- 13GPT-6 Luna (max)1209
- 14Gemini 3.1 Pro Grounding1207
- 15Grok 4.20 Multi-Agent1206
| # | ||||
|---|---|---|---|---|
| 1 |  GPT-5.6 Solxhigh | 1258 | 64.2 | 2026-10-13 |
| 2 | GPT-6.1 Solmax | 1250 | 67.3 | 2026-10-13 |
| 3 | GPT-6 Astramax | 1246 | 67.8 | 2026-10-13 |
| 4 |  Claude Fable 5.1high | 1233 | 64.0 | 2026-10-13 |
| 5 |  Qwen3.8 Maxmax | 1230 | 58.8 | 2026-10-13 |
| 6 | Claude Fable 5high | 1229 | 64.3 | 2026-10-13 |
| 7 | GPT-6 Solmax | 1228 | 63.3 | 2026-10-13 |
| 8 | GPT 5.5 Search | 1227 | – | 2026-10-13 |
| 9 | Claude Opus 4.6 Search | 1222 | – | 2026-10-13 |
| 10 | Claude Opus 5high | 1214 | 64.3 | 2026-10-13 |
| 11 | Claude Opus 5.5medium | 1211 | 62.0 | 2026-10-13 |
| 12 | Claude Opus 4.7 | 1210 | 63.1 | 2026-10-13 |
| 13 | GPT-6 Lunamax | 1209 | 56.7 | 2026-10-13 |
| 14 |  Gemini 3.1 Pro Grounding | 1207 | – | 2026-10-13 |
| 15 |  Grok 4.20 Multi-Agent | 1206 | 58.4 | 2026-10-13 |
| 16 | Grok 4.5 | 1206 | 59.8 | 2026-10-13 |
| 17 | Claude Sonnet 4.6 Search | 1202 | – | 2026-10-13 |
| 18 | Gemini 3 Pro Grounding | 1201 | – | 2026-10-13 |
| 19 | Grok 4.20 Beta1 | 1199 | – | 2026-10-13 |
| 20 | GPT 5.4 Search | 1196 | – | 2026-10-13 |
| 21 | o3 Search | 1196 | – | 2026-10-13 |
| 22 | Claude Opus 4.8 | 1195 | 60.0 | 2026-10-13 |
| 23 | Grok 4.1 Fast Search | 1195 | – | 2026-10-13 |
| 24 | Claude Sonnet 5 Search | 1193 | – | 2026-10-13 |
| 25 | Grok 4.3 | 1193 | 54.5 | 2026-10-13 |
| 26 |  ERNIE 5.1 | 1192 | – | 2026-10-13 |
| 27 | Gemini 3 Flash Grounding | 1189 | – | 2026-10-13 |
| 28 | GPT 5 Search | 1188 | – | 2026-10-13 |
| 29 | Claude Opus 4.5 Search | 1187 | – | 2026-10-13 |
| 30 | GPT 5.1 Search | 1184 | – | 2026-10-13 |
| 31 | Claude Sonnet 4.5 Search | 1176 | – | 2026-10-13 |
| 32 | Claude Opus 4.1 Search | 1167 | – | 2026-10-13 |
| 33 | GPT 5.2 Search | 1161 | – | 2026-10-13 |
| 34 | Grok 4 Fast Search | 1161 | – | 2026-10-13 |
| 35 | Claude Opus 4 Search | 1152 | – | 2026-10-13 |
| 36 | Gemini 2.5 Pro Grounding | 1144 | – | 2026-10-13 |
| 37 |  Ppl Sonar Prohigh | 1143 | – | 2026-10-13 |
| 38 | GPT 5.2 Searchno reasoning | 1136 | – | 2026-10-13 |
| 39 | Grok 4 Search | 1123 | – | 2026-10-13 |
| 40 | GPT 4o Search | 1083 | – | 2026-10-13 |
| 41 |  Diffbot Small Xl | 1063 | – | 2026-10-13 |
More agents & tools benchmarks · All LLM benchmarks
Related benchmarks
Top head-to-heads on Arena Search
Cite as: BenchLeader, “Arena Search leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/lmarena_search.
Arena Search: questions
- What does Arena Search measure?
- Search Arena compares models that answer with live web search, judged on the answer and its sourcing. Scores are reported in Elo-style ratings from pairwise comparisons; higher is better.
- What is the top score on LMArena Search?
- As of 6 Nov 2026, the highest score on Arena Search is 1258 by GPT-5.6 Sol (result dated 13 Oct 2026), ahead of GPT-6.1 Sol at 1250. Scores come from Arena (LMArena) and are re-read daily.
- Which AI model leads Arena Search?
- GPT-5.6 Sol leads Arena Search with 1258 as of 6 Nov 2026 (result dated 13 Oct 2026), ahead of GPT-6.1 Sol at 1250.
- How many models have Arena Search results?
- 41 model configurations have a Arena Search result on BenchLeader, all taken from Arena (LMArena).
- Who runs Arena Search and how often is it updated?
- Arena Search is published by Arena (LMArena). BenchLeader re-reads the published results once a day and records the date each result was published.
- Does Arena Search count toward the BenchLeader Index?
- No. Arena Search is shown for reference but left out of the composite index.