🏟️ Arena AI 排行榜

LMSYS 团队运营的真实人类盲投 ELO 榜 · 6 大维度 · 24,209,339.88 投票 · 自动 1 小时刷新

实时抓取缓存 1h

💬 Text Leaderboard · Top 30

通用文本对话 ELO 榜（盲投最多） · 390 模型（榜单共 390 名）·11,406,498 票

Rank	Model	Provider	Score (95% CI)	Votes	$ / M	Context
1	claude-fable-5 Anthropic	Anthropic	1507 ±5	21,439	$10 / $50	1M
2	claude-opus-4-6-high Anthropic	Anthropic	1505 ±4	72,466	$2.50 / $12.50	1M
3	claude-opus-4-7-high Anthropic	Anthropic	1502 ±4	60,268	$5 / $25	1M
4	muse-spark-1.2 (xHigh) Meta	Meta	1499 ±10	3,259	$1.25 / $4.25	N/A
5	claude-opus-4-6 Anthropic	Anthropic	1497 ±3	76,428	$2.50 / $12.50	1M
6	claude-opus-4-7 Anthropic	Anthropic	1494 ±4	61,369	$5 / $25	1M
7	claude-opus-5-high Anthropic	Anthropic	1494 ±5	19,792	$5 / $25	1M
8	claude-opus-5-max Anthropic	Anthropic	1491 ±7	9,559	$5 / $25	1M
9	Alibaba	qwen3.8-max	1491 ±8	6,908	$2 / $6	1M
10	muse-spark-1.1 Meta	Meta	1489 ±6	16,805	$1.25 / $4.25	N/A
11	Moonshot	kimi-k3-max	1489 ±6	11,864	N/A	N/A
12	muse-spark Meta	Meta	1488 ±6	13,596	N/A	N/A
13	Google	gemini-3.1-pro-preview	1486 ±3	94,989	$1 / $6	1M
14	Google	gemini-3-pro	1485 ±4	41,505	$2 / $12	1M
15	Google	gemini-3.6-flash-high	1484 ±6	13,693	$0.75 / $3.75	1M
16	claude-opus-4-8-high Anthropic	Anthropic	1482 ±5	40,588	$2.50 / $12.50	1M
17	OpenAI	gpt-5.5-high	1482 ±4	55,322	$2.50 / $15	1.1M
18	OpenAI	gpt-5.6-sol-xhigh	1481 ±6	15,450	$5 / $30	N/A
19	Google	gemini-3.5-flash-high	1477 ±5	25,746	$0.75 / $4.50	1M
20	OpenAI	gpt-5.5	1477 ±4	56,647	$2.50 / $15	1.1M
21	OpenAI	gpt-5.4-high	1477 ±4	60,819	$2.50 / $15	1.1M
22	OpenAI	gpt-5.2-chat-latest-20260210	1476 ±4	34,380	$1.75 / $14	128K
23	Google	gemini-3.5-flash-medium	1475 ±5	23,928	$0.75 / $4.50	1M
24	SpaceXAI	grok-4.20-beta1	1475 ±5	26,813	N/A	N/A
25	Alibaba	qwen3.7-max-preview	1474 ±10	3,718	$1.48 / $4.42	1M
26	claude-opus-4-8 Anthropic	Anthropic	1474 ±5	41,219	$2.50 / $12.50	1M
27	OpenAI	gpt-5.5-instant	1473 ±5	25,900	$2.50 / $15	1.1M
28	Google	gemini-3-flash	1473 ±4	30,859	$0.50 / $3	1M
29	claude-opus-4-5-20251101-high-32k Anthropic	Anthropic	1473 ±4	37,217	$5 / $25	200K
30	claude-sonnet-4-6 Anthropic	Anthropic	1472 ±4	66,602	$1.50 / $7.50	1M

关于数据：Arena.ai 排行榜基于人类对模型两两盲投结果，由 LMSYS Chatbot Arena 团队运营，是 LLM 评测领域公信力最高的 ELO 排名之一。所有分数均含 95% 置信区间，投票数越大分数越稳定。本页 实时抓取 arena.ai/leaderboard/ 5 个端点（text / vision / agent / search / text-to-image，webdev 已下线），缓存 1 小时。