Name: IBM 高难度提示词评测模型排行榜（2026-06-03） | AITNT
Creator: LM Arena
License: https://creativecommons.org/licenses/by/4.0/
Keywords: IBM,困难模式,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困难模式🌶️Hard Mode

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

10.5K

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
🥇	184 - 233	granite-4.1-8b IBM · Apache 2.0	1328±13.0	2.5K 票votes	1328.000 [1315.0, 1341.0]	49.00 · 不稳Volatile	—	—
🥈	218 - 251	ibm-granite-h-small IBM · Apache 2.0	1301±11.0	3.0K 票votes	1301.000 [1290.0, 1312.0]	33.00 · 不稳Volatile	N/A	N/A
🥉	267 - 298	granite-3.1-8b-instruct IBM · Apache 2.0	1230±20.0	798 票votes	1230.000 [1210.0, 1250.0]	31.00 · 不稳Volatile	N/A	N/A
4	271 - 302	granite-3.1-2b-instruct IBM · Apache 2.0	1218±19.0	832 票votes	1218.000 [1199.0, 1237.0]	31.00 · 不稳Volatile	N/A	N/A
5	283 - 313	granite-3.0-8b-instruct IBM · Apache 2.0	1202±15.0	1.7K 票votes	1202.000 [1187.0, 1217.0]	30.00 · 不稳Volatile	N/A	N/A
6	298 - 328	granite-3.0-2b-instruct IBM · Apache 2.0	1176±15.0	1.7K 票votes	1176.000 [1161.0, 1191.0]	30.00 · 不稳Volatile	N/A	N/A

🔍

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