Search Arena

View overall rankings across LLMs with integrated web search.

May 7, 2026

520,452 votes

29 models

Rank by

	Rank Spread
1	15	gpt-5.5-search OpenAI · Proprietary	1225±8	7,348	$5 / $30	1.1M
2	15	claude-opus-4-6-search Anthropic · Proprietary	1223±6	45,433	$5 / $25	1M
3	110	claude-opus-4-7 Anthropic · Proprietary	1213±8	7,482	$5 / $25	1M
4	19	gemini-3.1-pro-grounding Google · Proprietary	1212±7	25,762	N/A	N/A
5	110	grok-4.20-multi-agent-beta-0309 xAI · Proprietary	1211±6	25,220	$2 / $6	2M
6	313	grok-4.20-beta1 xAI · Proprietary	1203±6	45,659	N/A	N/A
7	313	gemini-3-pro-grounding Google · Proprietary	1202±5	37,291	$2 / $12	N/A
8	315	grok-4.3 xAI · Proprietary	1202±9	4,639	$1.25 / $2.50	1M
9	315	gpt-5.4-search OpenAI · Proprietary	1200±7	25,688	$2.50 / $15	1.1M
10	415	gemini-3-flash-grounding Google · Proprietary	1200±6	60,653	N/A	N/A
11	615	claude-sonnet-4-6-search Anthropic · Proprietary	1197±6	44,670	$3 / $15	1M
12	615	grok-4-1-fast-search xAI · Proprietary	1194±5	68,304	$0.20 / $0.50	2M
13	617	ernie-5.1 Baidu · Proprietary	1192±13	1,890	N/A	N/A
14	817	claude-opus-4-5-search Anthropic · Proprietary	1190±6	50,704	$5 / $25	200K
15	817	o3-search OpenAI · Proprietary	1189±7	20,706	$2 / $8	200K
16	1317	gpt-5.1-search OpenAI · Proprietary	1183±5	51,505	$1.25 / $10	400K
17	1318	gpt-5-search OpenAI · Proprietary	1180±8	20,842	$1.25 / $10	400K
18	1720	claude-sonnet-4-5-search Anthropic · Proprietary	1171±6	43,407	$3 / $15	1M
19	1821	gpt-5.2-search OpenAI · Proprietary	1166±6	45,162	$1.75 / $14	400K
20	1821	claude-opus-4-1-search Anthropic · Proprietary	1165±5	69,210	$15 / $75	200K
21	1922	grok-4-fast-search xAI · Proprietary	1160±5	42,962	$0.20 / $0.50	2M
22	2124	claude-opus-4-search Anthropic · Proprietary	1151±5	31,129	$15 / $75	200K
23	2226	gemini-2.5-pro-grounding Google · Proprietary	1143±5	74,491	$1.25 / $10	1M
24	2226	ppl-sonar-pro-high Perplexity · Proprietary	1142±6	28,624	$1 / $1	127.1K
25	2326	gpt-5.2-search-non-reasoning OpenAI · Proprietary	1139±6	62,223	$1.75 / $14	400K
26	2326	ppl-sonar-reasoning-pro-high Perplexity · Proprietary	1138±6	29,135	$1 / $1	127.1K
27	2727	grok-4-search xAI · Proprietary	1122±6	19,304	$3 / $15	256K
28	2829	api-gpt-4o-search OpenAI · Proprietary	1078±11	3,413	$30 / $60	8.2K
29	2829	diffbot-small-xl Diffbot · Apache 2.0	1062±8	6,407	N/A	N/A

Search Arena

With Style Control Leaderboard Plots

Confidence Intervals on Model Strength (via Bootstrapping)

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

Battle Count for Each Combination of Models (without Ties)

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