train-llm-from-scratch vs transformers
A neutral, constraint-first comparison - live GitHub stats and typed relationships, not marketing.
| train-llm-from-scratch | transformers | |
|---|---|---|
| Tagline | A method for training a Large Language Model (LLM) from scratch using Python. | 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models |
| Stars | 8.1k | 162k |
| Forks | 1.1k | 34k |
| Open issues | 2 | 2.5k |
| Language | Python | Python |
| License | MIT | Apache-2.0 |
| Last pushed | Jun 24, 2026 | Jul 7, 2026 |
| Categories | Model Training, LLM Frameworks | Model Training, Data & Retrieval, LLM Frameworks |
train-llm-from-scratch
Repository offering scripts to train LLMs using PyTorch, detailing the process from raw text to aligned model with every part implemented manually from basic neural networks through transformer architecture, up to supervised fine-tuning methods.
Python
transformers
Repo hosts a Python library and framework for NLP, text, audio, vision, multimodal AI model creation, training and inference using PyTorch.
Python