Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

WizardCoder-3B-V1.0

Model basics

Reasoning traces

Not supported

Context length

2K tokens

Max output length

No data

Model type

编程大模型

Release date

2023-08-27

Model file size

6.09GB

MoE architecture

Total params / Active params

30.0B / N/A

Knowledge cutoff

No data

Inference modes

No mode data

WizardCoder-3B-V1.0

Open source & experience

Code license

BigCode OpenRAIL-M v1

Weights license

BigCode OpenRAIL-M v1- 免费商用授权

GitHub repo

https://github.com/nlpxucan/WizardLM/tree/main/WizardCoder

Hugging Face

https://huggingface.co/WizardLM/WizardCoder-3B-V1.0

Live demo

No live demo

WizardCoder-3B-V1.0

Official resources

Paper

WizardCoder: Empowering Code Large Language Models with Evol-Instruct

DataLearnerAI blog

No blog post yet

WizardCoder-3B-V1.0

API details

API speed

No data

No public API pricing yet.

WizardCoder-3B-V1.0

Benchmark Results

No benchmark data to show.

WizardCoder-3B-V1.0

Publisher

WizardLM Team

View publisher details

WizardCoder-3B-V1.0

Model Overview

WizardCoder-3B-V1.0是由WizardLM发布的一系列编程大模型中的一个，基于StarCoder微调后得到，参数规模30亿，虽然参数不大，但是在HumanEval Pass@1上得分34.8，在MBPP @Pass1上得分37.4超过了很多更大规模的模型。

关于WizardCoder-3B-V1.0具体的编程排行和评测参考DataLearner大模型编程能力评测排行榜： https://www.datalearner.com/ai-models/llm-coding-evaluation

Foundation model

StarCoder

View details

DataLearner 官方微信

欢迎关注 DataLearner 官方微信，获得最新 AI 技术推送