jingyaogong/minimind
minimind
🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!
Builder
jingyaogong
jingyaogong • individual
Stars
45,460
Using upstream star count
Forks
5,536
Using upstream fork count
Open Issues
0
Activity Score
0/100
0 commits in 30d
Created
Jul 27, 2024
Project creation date
README Summary
MiniMind is a lightweight GPT implementation that enables training a 26-million parameter language model from scratch in just 2 hours. The project provides a complete pipeline for training small-scale GPT models with educational focus on understanding transformer architectures. It includes optimized training code, model architecture, and datasets specifically designed for quick experimentation and learning.
AI Dev Skills
Unmapped
Tags
Taxonomy
Deployment Context
Modalities
Skill Areas
Recent Activity
Updated 21 days ago
7 Days
0
30 Days
0
90 Days
0
Quality
research- Quality
- medium
- Maturity
- research
Categories
PM Skills
Languages
Timeline
- Project created
- Jul 27, 2024
- Forked
- Mar 23, 2026
- Your last push
- 21 days ago
- Upstream last push
- 9 days ago
- Tracked since
- Mar 23, 2026
Similar Repos
pgvector cosine similarity · $0
Loading…