Rankings

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

Search

Zoeken, grounding & RAG.

27 augustus 2026, 07:14

34 modellen

#	Model	Elo	±	Stemmen
	gpt-5.6-sol-xhigh OpenAI · Proprietary ±7 · 29.7k stemmen	1257	±7	29.7k
	claude-opus-4-6-search Anthropic · Proprietary ±5 · 134.7k stemmen	1253	±5	134.7k
	gpt-5.5-search OpenAI · Proprietary ±5 · 89.9k stemmen	1242	±5	89.9k
4	claude-opus-4-7 Anthropic · Proprietary ±5 · 91.4k stemmen	1233	±5	91.4k
5	claude-fable-5 Anthropic · Proprietary ±8 · 41.8k stemmen	1230	±8	41.8k
6	ernie-5.1 Baidu · Proprietary ±10 · 3.8k stemmen	1227	±10	3.8k
7	claude-sonnet-4-6-search Anthropic · Proprietary ±5 · 134.9k stemmen	1221	±5	134.9k
8	grok-4.5 SpaceXAI · Proprietary ±7 · 31.5k stemmen	1213	±7	31.5k
9	gemini-3.1-pro-grounding Google · Proprietary ±5 · 113.3k stemmen	1210	±5	113.3k
10	gemini-3-pro-grounding Google · Proprietary ±6 · 37.0k stemmen	1207	±6	37.0k
11	gpt-5.2-search OpenAI · Proprietary ±6 · 52.7k stemmen	1207	±6	52.7k
12	claude-opus-4-8 Anthropic · Proprietary ±6 · 71.0k stemmen	1204	±6	71.0k
13	grok-4.20-multi-agent-beta-0309 SpaceXAI · Proprietary ±5 · 109.6k stemmen	1204	±5	109.6k
14	gpt-5.1-search OpenAI · Proprietary ±5 · 59.9k stemmen	1199	±5	59.9k
15	gemini-3-flash-grounding Google · Proprietary ±5 · 149.3k stemmen	1198	±5	149.3k

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