GitHub - nlpxucan/WizardLM: LLMs build upon Evol Insturct: WizardLM, WizardCoder, WizardMath

Discover WizardLM - an innovative project enhancing large pre-trained language models with Evol Instruct technology. From coding to math tasks, WizardLM surpasses benchmarks like HumanEval. Engage with this academic research tool to boost AI intelligence! #AI #WizardLM 🧙‍♂️🔮

WizardLM is a project that empowers large pre-trained language models to follow complex instructions.
It includes different components like WizardCoder for coding tasks and WizardMath for mathematical reasoning.
WizardLM models have achieved high performance on various evaluation benchmarks like HumanEval and GSM8k.
The project uses innovative methods like Evol-Instruct to enhance the capabilities of language models.
WizardLM encourages feedback from users to improve its performance and address any weaknesses.
The resources associated with WizardLM are intended for academic research purposes only and not for commercial use.
The project provides detailed instructions on how to use the models for specific tasks like coding and mathematical reasoning.
WizardLM models have outperformed other state-of-the-art models like ChatGPT and Gemini Pro on different benchmarks.
WizardLM aims to create more intelligent AI by involving motivated students as interns.
The project emphasizes proper citation of papers if using data or code from WizardLM.