Search Arena

View rankings across LLMs with web search for real-time information, external knowledge, and grounded citations
Learn more about Search Arena at our blog

Last Updated

Jan 9, 2026

Total Votes

142,494

Total Models

15

/

/

Style Control

	Rank Spread
1	1◄─►3	gemini-3-pro-grounding	1214	±7	10,182	Google	Proprietary
2	1◄─►3	gpt-5.2-search	1211	±8	6,168	OpenAI	Proprietary
3	1◄─►3	gpt-5.1-search	1203	±8	8,059	OpenAI	Proprietary
4	4◄─►5	grok-4-1-fast-search	1177	±8	6,759	xAI	Proprietary
5	4◄─►5	grok-4-fast-search	1167	±5	25,077	xAI	Proprietary
6	6◄─►9	gemini-2.5-pro-grounding	1142	±5	29,975	Google	Proprietary
7	6◄─►9	o3-search	1140	±5	20,701	OpenAI	Proprietary
8	6◄─►12	grok-4-search	1138	±5	19,300	xAI	Proprietary
9	6◄─►12	ppl-sonar-reasoning-pro-high	1138	±5	29,657	Perplexity	Proprietary
10	8◄─►13	gpt-5-search	1130	±5	20,762	OpenAI	Proprietary
11	8◄─►13	claude-opus-4-1-search	1129	±5	29,488	Anthropic	Proprietary
12	8◄─►13	ppl-sonar-pro-high	1129	±5	29,018	Perplexity	Proprietary
13	10◄─►13	claude-opus-4-search	1126	±5	29,626	Anthropic	Proprietary
14	14◄─►15	diffbot-small-xl	1020	±8	6,390	Diffbot	Apache 2.0
15	14◄─►15	api-gpt-4o-search	1008	±10	3,451	OpenAI	Proprietary

Default Leaderboard Plots

Confidence Intervals on Model Strength (via Bootstrapping)

Battle Count for Each Combination of Models (without Ties)

Average Win Rate Against All Other Models (Uniform Sampling and No Ties)

Fraction of Model A Wins for All Non-tied A vs. B Battles