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

Dec 3, 2025

Total Votes

109,629

Total Models

13

/

/

Style Control

	Rank Spread (Upper-Lower)
1	1◄─►2	gemini-3-pro-grounding	1210	±11	2,537	Google	Proprietary
2	1◄─►2	gpt-5.1-search	1201	±12	2,430	OpenAI	Proprietary
3	3◄─►3	grok-4-fast-search	1164	±6	19,686	xAI	Proprietary
4	4◄─►7	gemini-2.5-pro-grounding	1142	±5	24,708	Google	Proprietary
5	4◄─►8	o3-search	1139	±5	21,009	OpenAI	Proprietary
6	4◄─►10	grok-4-search	1138	±5	19,302	xAI	Proprietary
7	4◄─►10	ppl-sonar-reasoning-pro-high	1138	±5	24,218	Perplexity	Proprietary
8	5◄─►11	claude-opus-4-1-search	1130	±5	24,186	Anthropic	Proprietary
9	6◄─►11	gpt-5-search	1130	±4	20,755	OpenAI	Proprietary
10	6◄─►11	claude-opus-4-search	1129	±5	24,390	Anthropic	Proprietary
11	8◄─►11	ppl-sonar-pro-high	1125	±5	23,882	Perplexity	Proprietary
12	12◄─►13	diffbot-small-xl	1020	±8	6,419	Diffbot	Apache 2.0
13	12◄─►13	api-gpt-4o-search	1008	±10	3,464	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