We introduced a new model designed for the Code generation task. Its test accuracy on the HumanEval base dataset surpasses that of GPT-4 Turbo (April 2024) and GPT-4o.
-
Updated
Jul 6, 2024 - Python
We introduced a new model designed for the Code generation task. Its test accuracy on the HumanEval base dataset surpasses that of GPT-4 Turbo (April 2024) and GPT-4o.
SkyCode是一个多语言开源编程大模型,采用GPT3模型结构,支持Java, JavaScript, C, C++, Python, Go, shell等多种主流编程语言,并能理解中文注释。模型可以对代码进行补全,拥有强大解题能力,使您从编程中解放出来,专心于解决更重要的问题。| SkyCode is an open source programming model, which adopts the GPT3 model structure. It supports Java, JavaScript, C, C++, Python, Go, shell and other languages, and can understand Chinese comments.
Code evaluation with *30-seconds-of-code* examples. Inspired by "Evaluating Large Language Models Trained on Code"
JetBrains Task: Leveraging software evolution data with LLMs
LLMs' performance analysis on CPU, GPU, Execution Time and Energy Usage
Add a description, image, and links to the humaneval topic page so that developers can more easily learn about it.
To associate your repository with the humaneval topic, visit your repo's landing page and select "manage topics."