BBH
更新于 2026-04-03
1,034 次浏览
问题数量
23
发布机构
Google
评测类别
综合评估
评测指标
Varies
支持语言
英语
难度等级
Expert
简介
BIG-Bench 的困难子集，包含更具挑战性的任务，用于评估模型的极限能力。
相关资源
BBH Model Score Leaderboard
Source: DataLearnerAI
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
模式说明:
normal
thinking
low
medium
high
deeper thinking
parallel_thinking
图表加载中...
Latest BBH model rankings and full benchmark leaderboard
Browse the latest scores, model modes, release dates, and parameter sizes for BBH.
Model release cutoff:
BBH详细排名数据表格
| 排名 | 模型 | |||
|---|---|---|---|---|
| 1 | ERNIE-4.5-300B-A47B Standard Mode | 94.30 | 2025-06-30 | 3000 |
| 2 | Claude 3.5 Sonnet New Standard Mode | 92.60 | 2024-10-22 | 未知 |
| 3 | DeepSeek-V3 Standard Mode | 92.30 | 2024-12-26 | 6810 |
| 4 | Hunyuan-TurboS Standard Mode | 92.20 | 2025-03-10 | 未知 |
| 5 | GPT-4o Standard Mode | 91.70 | 2024-05-13 | 未知 |
| 6 | Llama3.1-405B Instruct Standard Mode | 89.20 | 2024-07-23 | 4050 |
| 7 | Hunyuan-A13B-Instruct Standard Mode | 89.10 | 2025-06-27 | 800 |
| 8 | Qwen3-235B-A22B Standard Mode | 88.87 | 2025-04-28 | 2350 |
| 9 | Gemma 3 - 27B (IT) Standard Mode | 87.60 | 2025-03-12 | 270 |
| 10 | Qwen3-Next Standard Mode | 87.13 | 2025-09-11 | 800 |
| 11 | Qwen2.5-72B Standard Mode | 86.30 | 2024-09-18 | 727 |
| 12 | Gemma2-27B Standard Mode | 74.90 | 2024-05-14 | 270 |
| 13 | Gemma 2 - 9B Standard Mode | 68.20 | 2024-06-27 | 90 |
| 14 | Moonlight-16B-A3B-Instruct Standard Mode | 65.20 | 2025-02-23 | 160 |
| 15 | Llama3.1-8B Standard Mode | 57.70 | 2024-07-23 | 80 |
| 16 | Qwen2.5-3B Standard Mode | 56.30 | 2024-09-18 | 30 |
| 17 | Mistral-7B-Instruct-v0.3 Standard Mode | 56.10 | 2024-05-22 | 70 |
| 18 | Llama-3.2-3B Standard Mode | 46.80 | 2024-09-18 | 32 |
| 19 | Gemini 1.5 Pro Standard Mode | 0.00 | 2024-02-15 | 未知 |
| 20 | Amazon Nova Pro Standard Mode | 0.00 | 2024-12-03 | 未知 |