pavansai018/MayFei

MayFei is an open-source bilingual English-Chinese base language model trained from scratch on a single 8GB consumer GPU for general knowledge, text generation, reasoning, STEM, mathematics and code. It targets compute-efficient pretraining and reproducible performance against GPT-2 Small, GPT-2 Medium and Skywork-13B-Base across public benchmarks.

0

stars

45

commits

Python

primary language

Aug 1, 2026

updated

attention-is-all-you-need

attention-mechanism

autoregressive-models

decoder-only

generative-ai

llm

python

pytorch

transformer-architecture

transformers

README

Contributors

45 commits

Similar repos

zjunlp/knowlm-13b-base-v1.0

5

IDEA-CCNL/Fengshenbang-LM

Fengshenbang-LM(封神榜大模型)是IDEA研究院认知计算与自然语言研究中心主导的大模型开源体系，成为中文AIGC和认知智能的基础设施。

4,123

YeungNLP/firefly-ziya-13b

7

HqWu-HITCS/Awesome-Chinese-LLM

整理开源的中文大语言模型，以规模较小、可私有化部署、训练成本较低的模型为主，包括底座模型，垂直领域微调及应用，数据集与教程等。

22,757

AiHubCN/Awesome-Chinese-LLM

整理开源的中文大语言模型，以规模较小、可私有化部署、训练成本较低的模型为主，包括底座模型，垂直领域微调及应用，数据集与教程等。

22,757

FoundationVision/Liquid

(Accepted by IJCV) Liquid: Language Models are Scalable and Unified Multi-modal Generators

640

baichuan-inc/baichuan-7B

A large-scale 7B pretraining language model developed by BaiChuan-Inc.

5,649

baichuan-inc/Baichuan-13B-Base

187

Languages

Python

100.0%

pavansai018/MayFei

MayFei is an open-source bilingual English-Chinese base language model trained from scratch on a single 8GB consumer GPU for general knowledge, text generation, reasoning, STEM, mathematics and code. It targets compute-efficient pretraining and reproducible performance against GPT-2 Small, GPT-2 Medium and Skywork-13B-Base across public benchmarks.

0

stars

45

commits

Python

primary language

Aug 1, 2026

updated

attention-is-all-you-need

attention-mechanism

autoregressive-models

decoder-only

generative-ai

llm

python

pytorch

transformer-architecture

transformers

README

Similar repos

zjunlp/knowlm-13b-base-v1.0

5

IDEA-CCNL/Fengshenbang-LM

Fengshenbang-LM(封神榜大模型)是IDEA研究院认知计算与自然语言研究中心主导的大模型开源体系，成为中文AIGC和认知智能的基础设施。

4,123

YeungNLP/firefly-ziya-13b

7

HqWu-HITCS/Awesome-Chinese-LLM

整理开源的中文大语言模型，以规模较小、可私有化部署、训练成本较低的模型为主，包括底座模型，垂直领域微调及应用，数据集与教程等。

22,757

AiHubCN/Awesome-Chinese-LLM

整理开源的中文大语言模型，以规模较小、可私有化部署、训练成本较低的模型为主，包括底座模型，垂直领域微调及应用，数据集与教程等。

22,757

FoundationVision/Liquid

(Accepted by IJCV) Liquid: Language Models are Scalable and Unified Multi-modal Generators

640

baichuan-inc/baichuan-7B

A large-scale 7B pretraining language model developed by BaiChuan-Inc.

5,649

baichuan-inc/Baichuan-13B-Base

187

Contributors

45 commits

Languages

Python

100.0%