HumanEval
更新于 2026-04-11
3,887 次浏览
问题数量
164
发布机构
OpenAI
评测类别
编程与软件工程
评测指标
Accuracy
支持语言
英文
难度等级
基础
简介
一个包含 164 个手写编程问题的基准，用于评估模型生成代码的能力。
相关资源
HumanEval 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 HumanEval model rankings and full benchmark leaderboard
Browse the latest scores, model modes, release dates, and parameter sizes for HumanEval.
Model release cutoff:
HumanEval详细排名数据表格
| 排名 | 模型 | |||
|---|---|---|---|---|
| 1 | OpenAI o3-mini (high) Standard Mode | 97.60 | 2025-01-31 | Unknown |
| 2 | Claude Opus 4.6 Extended Thinking | 95.00 | 2026-02-05 | Unknown |
| 3 | Claude 3.5 Sonnet New Standard Mode | 93.70 | 2024-10-22 | Unknown |
| 4 | OpenAI o1-mini Standard Mode | 92.40 | 2024-09-12 | Unknown |
| 5 | Claude 3.5 Sonnet Standard Mode | 92.00 | 2024-06-21 | Unknown |
| 6 | Hunyuan-TurboS Standard Mode | 91.00 | 2025-03-10 | Unknown |
| 7 | GPT-4o(2024-11-20) Standard Mode | 90.20 | 2024-11-20 | Unknown |
| 8 | GPT-4o Standard Mode | 90.00 | 2024-05-13 | Unknown |
| 9 | Gemini 1.5 Pro Standard Mode | 89.00 | 2024-02-15 | Unknown |
| 10 | Llama3.1-405B Instruct Standard Mode | 89.00 | 2024-07-23 | 405B |
| 11 | Amazon Nova Pro Standard Mode | 89.00 | 2024-12-03 | Unknown |
| 12 | DeepSeek-V3 Standard Mode | 89.00 | 2024-12-26 | 681B |
| 13 | Mistral-Small-3.1-24B-Instruct-2503 Standard Mode | 88.41 | 2025-03-17 | 24B |
| 14 | Qwen2.5-32B Standard Mode | 88.40 | 2024-09-18 | 32B |
| 15 | Llama3.3-70B-Instruct Standard Mode | 88.40 | 2024-12-06 | 70B |
| 16 | Grok 2 Standard Mode | 88.40 | 2024-08-13 | 269B |
| 17 | Claude 3.5 Haiku Standard Mode | 88.10 | 2024-10-22 | Unknown |
| 18 | Gemma 3 - 27B (IT) Standard Mode | 87.80 | 2025-03-12 | 27B |
| 19 | GPT-4o mini Standard Mode | 87.20 | 2024-07-18 | Unknown |
| 20 | Codestral 25.01 Standard Mode | 86.60 | 2025-01-13 | Unknown |
| 21 | Claude3-Opus Standard Mode | 84.90 | 2024-03-04 | Unknown |
| 22 | Codestral Standard Mode | 81.10 | 2024-05-29 | 22B |
| 23 | Llama3.1-70B-Instruct Standard Mode | 80.50 | 2024-07-23 | 70B |
| 24 | Phi-4-mini-instruct (3.8B) Standard Mode | 74.40 | 2025-02-27 | 3.8B |
| 25 | Grok-1.5 Standard Mode | 74.10 | 2024-03-29 | Unknown |
| 26 | Qwen2.5-Max Standard Mode | 73.20 | 2025-01-28 | Unknown |
| 27 | GPT-4 Standard Mode | 67.00 | 2023-03-14 | 175B |
| 28 | Llama3.1-8B-Instruct Standard Mode | 66.50 | 2024-07-23 | 8B |
| 29 | C4AI Aya Vision 32B Standard Mode | 62.20 | 2025-03-04 | 32B |
| 30 | Qwen2.5-72B Standard Mode | 59.10 | 2024-09-18 | 72.7B |
滚动或悬停加载剩余 9 条