Skip to content
Bencher

TestsBenchmarks · Human preference

LMArena Search Arena

Crowd-voted Elo (raw) for search-grounded answers from models with web search/grounding enabled.

Elo ratingHigher is betterThe test's websiteOfficial page ↗

The best 15 · 18 models tested

Top 15 · best setting per model · 18 models, 18 results (18 independent)

Measured by independent testersIndependent Reported by the makerVendor-reported
LMArena Search Arena leaderboard
Thinking levelSetting
1GPT-5.6 SolOpenAI1257Extra highgpt-5.6-sol-xhighIndependent testIndependentLMArena ↗—24 Aug 2026
2Claude Opus 4.6Anthropic1253Defaultclaude-opus-4-6-searchIndependent testIndependentLMArena ↗—24 Aug 2026
3GPT-5.5OpenAI1242Defaultgpt-5.5-searchIndependent testIndependentLMArena ↗—24 Aug 2026
4Claude Opus 4.7Anthropic1233Defaultclaude-opus-4-7Independent testIndependentLMArena ↗—24 Aug 2026
5Claude Fable 5Anthropic1230Highclaude-fable-5-highIndependent testIndependentLMArena ↗—24 Aug 2026
6Claude Sonnet 4.6Anthropic1221Defaultclaude-sonnet-4-6-searchIndependent testIndependentLMArena ↗—24 Aug 2026
7Grok 4.5SpaceXAI (formerly xAI)1213Defaultgrok-4.5Independent testIndependentLMArena ↗—24 Aug 2026
8Gemini 3.1 Pro (Preview)Google (Gemini / DeepMind)1211Defaultgemini-3.1-pro-groundingIndependent testIndependentLMArena ↗—24 Aug 2026
9Gemini 3 ProGoogle (Gemini / DeepMind)1207Defaultgemini-3-pro-groundingIndependent testIndependentLMArena ↗—24 Aug 2026
10GPT-5.2OpenAI1207Defaultgpt-5.2-searchIndependent testIndependentLMArena ↗—24 Aug 2026
11Claude Opus 4.8Anthropic1204Defaultclaude-opus-4-8Independent testIndependentLMArena ↗—24 Aug 2026
12Grok 4.20 Multi-AgentSpaceXAI (formerly xAI)1204Defaultgrok-4.20-multi-agent-beta-0309Independent testIndependentLMArena ↗—24 Aug 2026
13GPT-5.1OpenAI1199Defaultgpt-5.1-searchIndependent testIndependentLMArena ↗—24 Aug 2026
14Gemini 3 Flash PreviewGoogle (Gemini / DeepMind)1198Defaultgemini-3-flash-groundingIndependent testIndependentLMArena ↗—24 Aug 2026
15GPT-5.4OpenAI1197Defaultgpt-5.4-searchIndependent testIndependentLMArena ↗—24 Aug 2026
16Claude Sonnet 5Anthropic1194Defaultclaude-sonnet-5-searchIndependent testIndependentLMArena ↗—24 Aug 2026
17Grok 4.20SpaceXAI (formerly xAI)1189Defaultgrok-4.20-beta1Independent testIndependentLMArena ↗—24 Aug 2026
18Claude Opus 4.5Anthropic1180Defaultclaude-opus-4-5-searchIndependent testIndependentLMArena ↗—24 Aug 2026