AI Search Leaderboard: Top Models for Research

Search Arena

View overall rankings across LLMs with integrated web search.

Aug 24, 2026

1,110,523 votes

34 models

	Rank Spread
1	12	gpt-5.6-sol-xhigh OpenAI · Proprietary	1257±7	29,663	$4 / $20	1.1M
2	12	claude-opus-4-6-search Anthropic · Proprietary	1253±5	134,699	$5 / $25	1M
3	35	gpt-5.5-search OpenAI · Proprietary	1242±5	89,873	$5 / $30	1.1M
4	36	claude-opus-4-7 Anthropic · Proprietary	1233±5	91,394	$5 / $25	1M
5	37	claude-fable-5-high Anthropic · Proprietary	1230±8	41,795	$10 / $50	1M
6	48	ernie-5.1 Baidu · Proprietary	1227±10	3,788	$0.56 / $2.54	119K
7	58	claude-sonnet-4-6-search Anthropic · Proprietary	1221±5	134,905	$1.50 / $7.50	1M
8	613	grok-4.5 SpaceXAI · Proprietary	1213±7	31,505	$2 / $6	500K
9	813	gemini-3.1-pro-grounding Google · Proprietary	1210±5	113,282	N/A	N/A
10	816	gemini-3-pro-grounding Google · Proprietary	1207±6	37,024	$2 / $12	N/A
11	816	gpt-5.2-search OpenAI · Proprietary	1207±6	52,712	$0.88 / $7	400K
12	817	claude-opus-4-8 Anthropic · Proprietary	1204±6	70,998	$5 / $25	1M
13	817	grok-4.20-multi-agent-beta-0309 SpaceXAI · Proprietary	1204±5	109,553	$1.25 / $2.50	1M
14	1018	gpt-5.1-search OpenAI · Proprietary	1199±5	59,909	$0.63 / $5	400K
15	1018	gemini-3-flash-grounding Google · Proprietary	1198±5	149,334	N/A	N/A
16	1018	gpt-5.4-search OpenAI · Proprietary	1197±5	110,116	$2.50 / $15	1.1M
17	1218	claude-sonnet-5-search Anthropic · Proprietary	1194±7	40,230	$2 / $10	1M
18	1419	grok-4.20-beta1 SpaceXAI · Proprietary	1189±6	53,921	N/A	N/A
19	1822	claude-opus-4-5-search Anthropic · Proprietary	1180±6	61,573	$5 / $25	200K
20	1923	gpt-5.2-search-non-reasoning OpenAI · Proprietary	1172±6	75,658	$0.88 / $7	400K
21	1923	grok-4-1-fast-search SpaceXAI · Proprietary	1171±5	81,507	$1.25 / $2.50	2M
22	1923	grok-4-fast-search SpaceXAI · Proprietary	1171±5	41,794	$0.20 / $0.50	2M
23	2024	grok-4.3 SpaceXAI · Proprietary	1165±5	91,483	$1.25 / $2.50	1M
24	2324	claude-sonnet-4-5-search Anthropic · Proprietary	1158±5	127,378	$3 / $15	1M
25	2529	claude-opus-4-1-search Anthropic · Proprietary	1148±5	76,933	$15 / $75	200K
26	2529	o3-search OpenAI · Proprietary	1144±5	20,644	$1 / $4	200K
27	2530	gemini-2.5-pro-grounding Google · Proprietary	1142±5	83,404	$0.63 / $5	1M
28	2531	grok-4-search SpaceXAI · Proprietary	1142±6	19,108	$3 / $15	N/A
29	2531	ppl-sonar-reasoning-pro-high Perplexity AI · Proprietary	1139±6	29,055	$1 / $1	127.1K
30	2732	gpt-5-search OpenAI · Proprietary	1133±6	20,781	$0.63 / $5	400K
31	2832	ppl-sonar-pro-high Perplexity AI · Proprietary	1130±6	28,519	$1 / $1	127.1K
32	3032	claude-opus-4-search Anthropic · Proprietary	1126±5	31,037	$15 / $75	200K
33	3334	diffbot-small-xl Diffbot · Apache 2.0	1023±8	6,388	N/A	N/A
34	3334	api-gpt-4o-search OpenAI · Proprietary	1006±11	3,411	$30 / $60	8.2K

Search Arena

View overall rankings across LLMs with integrated web search.

Default Leaderboard Plots

Fraction of Model A Wins in A vs. B Battles (Ties Excluded)

Average Win Rate Against Each Opponent (Ties Excluded)

Battle Count for Each Pair of Models (Ties Excluded)

Confidence Intervals on Model Strength (via Bootstrapping)