Name: Anthropic 多轮对话模型排行榜（2026-08-22） | AITNT
Creator: LM Arena
License: https://creativecommons.org/licenses/by/4.0/
Keywords: Anthropic,多轮对话,AI 大语言模型排行榜,AI 模型排行

LM Arena 大模型全球实力榜LM Arena Global AI Leaderboard多轮对话💬Multi-turn

查看各种 AI 大语言模型在数学推理、代码生成、创意写作及其他开放式文本任务中的综合 Elo 积分排名，数据来源于真实用户的匿名盲测投票。

2026-08-22

▾

上榜模型Models

参赛厂商Providers

333.3K

总投票数Total Votes

📖 指标说明 — 读懂这张榜单📖 Metrics Guide — Understanding the Leaderboard

🎯

榜单主指标（Score）Primary Metric (Score)

每个榜单会采用官网对应的核心排序指标。通用 Arena 榜单通常是模型竞技积分；智能体榜单则展示净提升及多项任务指标。

Each leaderboard uses the official primary ranking metric. Most Arena boards use competitive scores; the Agent board uses net improvement and task-specific metrics.

🔢

精确值（Float Value）Float Value

主指标的完整浮点数值。对于通用 Arena 榜单可用于区分整数分相近的模型；智能体榜单以百分比和置信区间展示。

The full floating-point value of the primary metric. It helps distinguish close scores on standard boards; Agent metrics are shown as percentages with confidence intervals.

📊

方差（Variance）Variance

衡量结果的统计稳定程度。绿色·稳定（<5）更可信；橙色·波动（5~12）；红色·不稳（>12）说明排名还可能明显变化。

Measures statistical stability. Green·Stable (<5); Orange·Unstable (5–12); Red·Volatile (>12) means the ranking may still move significantly.

↔️

置信区间（95% CI = ±N）95% Confidence Interval (CI = ±N)

95% 置信区间反映当前估计的不确定范围，显示为 ±N。在相同计算口径下，N 越小说明估计越集中、排名越稳定。

The 95% confidence interval, shown as ±N, summarizes uncertainty in the current estimate. Under the same methodology, a smaller N indicates a more precise estimate and a more stable ranking.

🗳️

投票数（Vote Count）Vote Count

该模型或厂商参与评测的总次数。数量越高，结果的统计置信度越高、越不容易受单次样本影响而波动。

Total evaluation volume for the model or lab. Higher counts mean higher statistical confidence and less sensitivity to individual samples.

📜

开源协议（License）License

Apache/Llama：完全开源可商用；MIT：开源限制极少；Proprietary：闭源商业模型。协议决定你能否把它集成到自己的产品里。

Apache/Llama: Fully open-source, commercial use allowed; MIT: Open-source, minimal restrictions; Proprietary: Closed-source commercial model. License determines if you can integrate it into your product.

排名Rank	名次区间Rank Range	模型Model		投票数Votes	精确分 & 区间Float Score & CI	方差Variance	价格 ($/1M)Price ($/1M)	上下文Context
🥇	1 - 9	claude-opus-4-6-high Anthropic · Proprietary	1517±7.0	12.4K 票votes	1517.000 [1510.0, 1524.0]	8.00 · 波动Unstable	—	—
🥈	1 - 10	claude-opus-4-7-high Anthropic · Proprietary	1516±7.0	10.6K 票votes	1516.000 [1509.0, 1523.0]	9.00 · 波动Unstable	—	—
🥉	1 - 12	claude-fable-5 Anthropic · Proprietary	1516±10.0	4.1K 票votes	1516.000 [1506.0, 1526.0]	11.00 · 波动Unstable	—	—
4	1 - 10	claude-opus-4-7 Anthropic · Proprietary	1515±7.0	10.9K 票votes	1515.000 [1508.0, 1522.0]	9.00 · 波动Unstable	$5 / $25	1M
5	1 - 12	claude-opus-4-6 Anthropic · Proprietary	1512±6.0	13.6K 票votes	1512.000 [1506.0, 1518.0]	11.00 · 波动Unstable	$5 / $25	1M
6	4 - 29	claude-opus-4-8-high Anthropic · Proprietary	1499±8.0	8.0K 票votes	1499.000 [1491.0, 1507.0]	25.00 · 不稳Volatile	—	—
7	6 - 31	claude-opus-4-8 Anthropic · Proprietary	1497±8.0	8.2K 票votes	1497.000 [1489.0, 1505.0]	25.00 · 不稳Volatile	—	—
8	6 - 55	claude-opus-5-high Anthropic · Proprietary	1488±10.0	4.4K 票votes	1488.000 [1478.0, 1498.0]	49.00 · 不稳Volatile	—	—
9	6 - 52	claude-opus-4-5-20251101-high-32k Anthropic · Proprietary	1487±8.0	6.5K 票votes	1487.000 [1479.0, 1495.0]	46.00 · 不稳Volatile	—	—
10	8 - 55	claude-opus-4-5-20251101 Anthropic · Proprietary	1484±6.0	12.7K 票votes	1484.000 [1478.0, 1490.0]	47.00 · 不稳Volatile	$5 / $25	200K
11	6 - 67	claude-opus-5-max Anthropic · Proprietary	1482±14.0	2.1K 票votes	1482.000 [1468.0, 1496.0]	61.00 · 不稳Volatile	—	—
12	12 - 62	claude-sonnet-4-6 Anthropic · Proprietary	1481±7.0	12.0K 票votes	1481.000 [1474.0, 1488.0]	50.00 · 不稳Volatile	$3 / $15	1M
13	15 - 63	claude-sonnet-4-5-20250929 Anthropic · Proprietary	1479±6.0	14.2K 票votes	1479.000 [1473.0, 1485.0]	48.00 · 不稳Volatile	$3 / $15	200K
14	19 - 70	claude-opus-4-1-20250805-thinking-16k Anthropic · Proprietary	1472±7.0	8.6K 票votes	1472.000 [1465.0, 1479.0]	51.00 · 不稳Volatile	$15 / $75	200K
15	16 - 74	claude-sonnet-5-high Anthropic · Proprietary	1471±9.0	4.8K 票votes	1471.000 [1462.0, 1480.0]	58.00 · 不稳Volatile	—	—
16	24 - 71	claude-sonnet-4-5-20250929-high-32k Anthropic · Proprietary	1470±6.0	14.1K 票votes	1470.000 [1464.0, 1476.0]	47.00 · 不稳Volatile	—	—
17	28 - 71	claude-opus-4-1-20250805 Anthropic · Proprietary	1469±6.0	13.4K 票votes	1469.000 [1463.0, 1475.0]	43.00 · 不稳Volatile	$15 / $75	200K
18	69 - 119	claude-opus-4-20250514-thinking-16k Anthropic · Proprietary	1437±8.0	6.3K 票votes	1437.000 [1429.0, 1445.0]	50.00 · 不稳Volatile	$15 / $75	200K
19	84 - 130	claude-opus-4-20250514 Anthropic · Proprietary	1428±8.0	7.8K 票votes	1428.000 [1420.0, 1436.0]	46.00 · 不稳Volatile	$15 / $75	200K
20	90 - 131	claude-haiku-4-5-20251001 Anthropic · Proprietary	1425±5.0	21.8K 票votes	1425.000 [1420.0, 1430.0]	41.00 · 不稳Volatile	$1 / $5	200K
21	93 - 146	claude-sonnet-4-20250514-thinking-32k Anthropic · Proprietary	1420±8.0	6.0K 票votes	1420.000 [1412.0, 1428.0]	53.00 · 不稳Volatile	$3 / $15	1M
22	108 - 162	claude-sonnet-4-20250514 Anthropic · Proprietary	1408±8.0	7.2K 票votes	1408.000 [1400.0, 1416.0]	54.00 · 不稳Volatile	$3 / $15	1M
23	122 - 169	claude-3-7-sonnet-20250219-thinking-32k Anthropic · Proprietary	1398±8.0	6.4K 票votes	1398.000 [1390.0, 1406.0]	47.00 · 不稳Volatile	$3 / $15	200K
24	124 - 169	claude-3-7-sonnet-20250219 Anthropic · Proprietary	1396±8.0	7.1K 票votes	1396.000 [1388.0, 1404.0]	45.00 · 不稳Volatile	$3 / $15	200K
25	137 - 173	claude-3-5-sonnet-20241022 Anthropic · Proprietary	1389±6.0	16.1K 票votes	1389.000 [1383.0, 1395.0]	36.00 · 不稳Volatile	$3 / $15	200K
26	179 - 214	claude-3-5-sonnet-20240620 Anthropic · Proprietary	1349±7.0	14.9K 票votes	1349.000 [1342.0, 1356.0]	35.00 · 不稳Volatile	$3 / $15	200K
27	199 - 244	claude-3-5-haiku-20241022 Anthropic · Proprietary	1327±6.0	11.5K 票votes	1327.000 [1321.0, 1333.0]	45.00 · 不稳Volatile	$0.80 / $4	200K
28	201 - 249	claude-3-opus-20240229 Anthropic · Proprietary	1321±6.0	31.1K 票votes	1321.000 [1315.0, 1327.0]	48.00 · 不稳Volatile	$15 / $75	200K
29	251 - 284	claude-3-sonnet-20240229 Anthropic · Proprietary	1278±8.0	17.1K 票votes	1278.000 [1270.0, 1286.0]	33.00 · 不稳Volatile	$3 / $15	200K
30	279 - 297	claude-3-haiku-20240307 Anthropic · Proprietary	1244±7.0	19.3K 票votes	1244.000 [1237.0, 1251.0]	18.00 · 不稳Volatile	$0.25 / $1.25	200K

🔍

没有找到匹配的模型，换个关键词试试？No matching models found. Try a different keyword?