WizardCoder: Empowering Code Large Language Models with Evol-Instruct

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Luo, Ziyang, Xu, Can, Zhao, Pu, Sun, Qingfeng, Geng, Xiubo, Hu, Wenxiang, Tao, Chongyang, Ma, Jing, Lin, Qingwei, Jiang, Daxin
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915306299981824
author Luo, Ziyang
Xu, Can
Zhao, Pu
Sun, Qingfeng
Geng, Xiubo
Hu, Wenxiang
Tao, Chongyang
Ma, Jing
Lin, Qingwei
Jiang, Daxin
author_facet Luo, Ziyang
Xu, Can
Zhao, Pu
Sun, Qingfeng
Geng, Xiubo
Hu, Wenxiang
Tao, Chongyang
Ma, Jing
Lin, Qingwei
Jiang, Daxin
contents Code Large Language Models (Code LLMs), such as StarCoder, have demonstrated exceptional performance in code-related tasks. However, most existing models are solely pre-trained on extensive raw code data without instruction fine-tuning. In this paper, we introduce WizardCoder, which empowers Code LLMs with complex instruction fine-tuning, by adapting the Evol-Instruct method to the domain of code. Through comprehensive experiments on four prominent code generation benchmarks, namely HumanEval, HumanEval+, MBPP, and DS-1000, we unveil the exceptional capabilities of our model. It surpasses all other open-source Code LLMs by a substantial margin. Moreover, our model even outperforms the largest closed LLMs, Anthropic's Claude and Google's Bard, on HumanEval and HumanEval+. Our code, model weights, and data are public at https://github.com/nlpxucan/WizardLM
format Preprint
id arxiv_https___arxiv_org_abs_2306_08568
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle WizardCoder: Empowering Code Large Language Models with Evol-Instruct
Luo, Ziyang
Xu, Can
Zhao, Pu
Sun, Qingfeng
Geng, Xiubo
Hu, Wenxiang
Tao, Chongyang
Ma, Jing
Lin, Qingwei
Jiang, Daxin
Computation and Language
Artificial Intelligence
Code Large Language Models (Code LLMs), such as StarCoder, have demonstrated exceptional performance in code-related tasks. However, most existing models are solely pre-trained on extensive raw code data without instruction fine-tuning. In this paper, we introduce WizardCoder, which empowers Code LLMs with complex instruction fine-tuning, by adapting the Evol-Instruct method to the domain of code. Through comprehensive experiments on four prominent code generation benchmarks, namely HumanEval, HumanEval+, MBPP, and DS-1000, we unveil the exceptional capabilities of our model. It surpasses all other open-source Code LLMs by a substantial margin. Moreover, our model even outperforms the largest closed LLMs, Anthropic's Claude and Google's Bard, on HumanEval and HumanEval+. Our code, model weights, and data are public at https://github.com/nlpxucan/WizardLM
title WizardCoder: Empowering Code Large Language Models with Evol-Instruct
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2306.08568