BenchMark·Hub

HumanEval+

A已审核Variant
代码代码生成HumanEval 家族开放

发布方:EvalPlus

热度9.1▲ 0.3

HumanEval+

evalplus / pass@1 · 125 行结果

排名模型子项分数Metric时间
#1O1 Preview (Sept 2024)
HumanEval+
evalplus
89%pass@12026-08-28
#2O1 Mini (Sept 2024)
HumanEval+
evalplus
89%pass@12026-08-28
#3Qwen2.5-Coder-32B-Instruct
Qwen
HumanEval+
evalplus
87.2%pass@12026-08-28
#4GPT 4o (Aug 2024)
HumanEval+
evalplus
87.2%pass@12026-08-28
#5DeepSeek-V3 (Nov 2024)
deepseek-ai
HumanEval+
evalplus
86.6%pass@12026-08-28
#6GPT-4-Turbo (April 2024)
HumanEval+
evalplus
86.6%pass@12026-08-28
#7DeepSeek-V2.5 (Nov 2024)
deepseek-ai
HumanEval+
evalplus
83.5%pass@12026-08-28
#8GPT 4o Mini (July 2024)
HumanEval+
evalplus
83.5%pass@12026-08-28
#9DeepSeek-Coder-V2-Instruct
deepseek-ai
HumanEval+
evalplus
82.3%pass@12026-08-28
#10Claude Sonnet 3.5 (June 2024)
HumanEval+
evalplus
81.7%pass@12026-08-28
#11GPT-4-Turbo (Nov 2023)
HumanEval+
evalplus
81.7%pass@12026-08-28
#12Grok Beta
HumanEval+
evalplus
80.5%pass@12026-08-28
#13Gemini 1.5 Pro 002
HumanEval+
evalplus
79.3%pass@12026-08-28
#14GPT-4 (May 2023)
HumanEval+
evalplus
79.3%pass@12026-08-28
#15CodeQwen1.5-7B-Chat
Qwen
HumanEval+
evalplus
78.7%pass@12026-08-28
#16OpenCoder-8B-Instruct
infly
HumanEval+
evalplus
77.4%pass@12026-08-28
#17claude-3-opus (Mar 2024)
HumanEval+
evalplus
77.4%pass@12026-08-28
#18Gemini 1.5 Flash 002
HumanEval+
evalplus
75.6%pass@12026-08-28
#19DeepSeek-Coder-33B-instruct
deepseek-ai
HumanEval+
evalplus
75%pass@12026-08-28
#20Codestral-22B-v0.1
mistralai
HumanEval+
evalplus
73.8%pass@12026-08-28
#21OpenCodeInterpreter-DS-33B
m-a-p
HumanEval+
evalplus
73.8%pass@12026-08-28
#22WizardCoder-33B-V1.1
WizardLM
HumanEval+
evalplus
73.2%pass@12026-08-28
#23Artigenz-Coder-DS-6.7B
Artigenz
HumanEval+
evalplus
72.6%pass@12026-08-28
#24Llama3-70B-instruct
meta-llama
HumanEval+
evalplus
72%pass@12026-08-28
#25Mixtral-8x22B-Instruct-v0.1
mistralai
HumanEval+
evalplus
72%pass@12026-08-28
125 条结果 · 当前 1-25
1 / 5