Name: Google 作业解答模型排行榜（2026-06-03） | AITNT
Creator: LM Arena
License: https://creativecommons.org/licenses/by/4.0/
Keywords: Google,作业解答,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作业解答📚Homework

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。

2026-06-03

▾

上榜模型Models

参赛厂商Providers

37.7K

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 - 20	gemini-3-flash Google · Proprietary	1313±10.0	3.9K 票votes	1313.000 [1303.0, 1323.0]	19.00 · 不稳Volatile	$0.50 / $3	1M
🥈	1 - 24	gemini-3-pro Google · Proprietary	1310±14.0	1.8K 票votes	1310.000 [1296.0, 1324.0]	23.00 · 不稳Volatile	$2 / $12	1M
🥉	2 - 22	gemini-3.1-pro-preview Google · Proprietary	1310±11.0	3.6K 票votes	1310.000 [1299.0, 1321.0]	20.00 · 不稳Volatile	$2 / $12	1M
4	5 - 29	gemma-4-31b Google · Apache 2.0	1297±10.0	4.4K 票votes	1297.000 [1287.0, 1307.0]	24.00 · 不稳Volatile	$0.14 / $0.40	262.1K
5	7 - 31	gemini-3-flash (thinking-minimal) Google · Proprietary	1294±11.0	3.7K 票votes	1294.000 [1283.0, 1305.0]	24.00 · 不稳Volatile	$0.50 / $3	1M
6	11 - 40	gemma-4-26b-a4b Google · Apache 2.0	1284±12.0	2.7K 票votes	1284.000 [1272.0, 1296.0]	29.00 · 不稳Volatile	N/A	N/A
7	15 - 49	gemini-3.1-flash-lite-preview Google · Proprietary	1272±12.0	3.0K 票votes	1272.000 [1260.0, 1284.0]	34.00 · 不稳Volatile	$0.25 / $1.50	1M
8	18 - 47	gemini-2.5-pro Google · Proprietary	1272±9.0	5.4K 票votes	1272.000 [1263.0, 1281.0]	29.00 · 不稳Volatile	$1.25 / $10	1M
9	13 - 59	gemini-2.5-flash-preview-09-2025 Google · Proprietary	1266±22.0	624 票votes	1266.000 [1244.0, 1288.0]	46.00 · 不稳Volatile	$0.30 / $2.50	1M
10	33 - 61	gemini-2.5-flash Google · Proprietary	1237±9.0	4.9K 票votes	1237.000 [1228.0, 1246.0]	28.00 · 不稳Volatile	$0.30 / $2.50	1M
11	50 - 78	gemini-2.5-flash-lite-preview-09-2025-no-thinking Google · Proprietary	1196±22.0	621 票votes	1196.000 [1174.0, 1218.0]	28.00 · 不稳Volatile	$0.10 / $0.40	1M
12	53 - 80	gemini-2.5-flash-lite-preview-06-17-thinking Google · Proprietary	1188±16.0	1.4K 票votes	1188.000 [1172.0, 1204.0]	27.00 · 不稳Volatile	$0.10 / $0.40	1M
13	53 - 80	gemini-2.0-flash-001 Google · Proprietary	1187±21.0	789 票votes	1187.000 [1166.0, 1208.0]	27.00 · 不稳Volatile	$0.10 / $0.40	1M
14	58 - 80	gemma-3-27b-it Google · Gemma	1169±18.0	1.1K 票votes	1169.000 [1151.0, 1187.0]	22.00 · 不稳Volatile	$0.08 / $0.16	131.1K

🔍

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