
MiniMind
github.com/jingyaogong/minimind- Category
- AI Tools
- Rank
- No. 135Tools index
- Pricing
- Open Source
- Type
- TOOL
- Builder
- jingyaogong
- GitHub
- 54.5k stars
- Latest release
- v2
- Added
- Apr 14, 2026
About
A lightweight 64M parameter GPT model that can be trained from scratch in just 2 hours on a single RTX 3090. Provides complete training pipeline including pretraining, SFT, LoRA, RLHF, and tool use capabilities.
What it can do
Train a GPT language model from scratch
Training dataset and configuration parameters → 64M parameter trained language model
Perform supervised fine-tuning on pre-trained model
Pre-trained model and supervised training dataset → Fine-tuned language model
Apply LoRA (Low-Rank Adaptation) fine-tuning
Base model and LoRA training data → LoRA-adapted model with efficient parameter updates
Execute RLHF (Reinforcement Learning from Human Feedback) training
Model and human feedback data → RLHF-optimized language model
Enable tool use capabilities in language model
Model and tool integration configuration → Language model with tool calling abilities
Run complete training pipeline on single GPU
RTX 3090 GPU and training configuration → Fully trained model in approximately 2 hours
Why it made the leaderboard
MiniMind fills a crucial gap in AI education by making LLM training genuinely accessible to individual developers. Unlike research projects requiring massive resources, this provides a complete, runnable pipeline in just 2 hours on consumer hardware. The 44k GitHub stars and focus on educational value demonstrate real impact in democratizing AI knowledge.
Tags
Tech Stack
Comments (0)
No comments yet
Indexed by a proprietary survey. Corrections welcome.