Rankings

Arena.ai leaderboards: Elo op basis van blind head-to-head stemmen. Kies een categorie.

Search

Zoeken, grounding & RAG.

28 mei 2026, 19:45

29 modellen

#	Model	Elo	±	Stemmen
	claude-opus-4-6-search Anthropic · Proprietary ±6 · 48.7k stemmen	1251	±6	48.7k
	gpt-5.5-search OpenAI · Proprietary ±8 · 9.5k stemmen	1239	±8	9.5k
	claude-opus-4-7 Anthropic · Proprietary ±8 · 9.7k stemmen	1237	±8	9.7k
4	ernie-5.1 Baidu · Proprietary ±12 · 2.3k stemmen	1226	±12	2.3k
5	claude-sonnet-4-6-search Anthropic · Proprietary ±6 · 48.1k stemmen	1219	±6	48.1k
6	gemini-3.1-pro-grounding Google · Proprietary ±7 · 28.0k stemmen	1216	±7	28.0k
7	gpt-5.2-search OpenAI · Proprietary ±6 · 47.1k stemmen	1210	±6	47.1k
8	grok-4.20-multi-agent-beta-0309 xAI · Proprietary ±7 · 27.5k stemmen	1209	±7	27.5k
9	gemini-3-pro-grounding Google · Proprietary ±5 · 37.3k stemmen	1208	±5	37.3k
10	gemini-3-flash-grounding Google · Proprietary ±6 · 62.9k stemmen	1206	±6	62.9k
11	gpt-5.1-search OpenAI · Proprietary ±6 · 53.7k stemmen	1199	±6	53.7k
12	gpt-5.4-search OpenAI · Proprietary ±7 · 27.9k stemmen	1199	±7	27.9k
13	grok-4.20-beta1 xAI · Proprietary ±6 · 49.0k stemmen	1193	±6	49.0k
14	grok-4.3 xAI · Proprietary ±8 · 6.9k stemmen	1189	±8	6.9k
15	claude-opus-4-5-search Anthropic · Proprietary ±6 · 53.4k stemmen	1182	±6	53.4k

Elo-score op basis van blind head-to-head stemmen. Hoger is beter. ± is het 95% betrouwbaarheidsinterval. Zelfde formaat als Arena.ai.