LLMs-from-scratch vs recipes
A neutral, constraint-first comparison - live GitHub stats and typed relationships, not marketing.
| LLMs-from-scratch | recipes | |
|---|---|---|
| Tagline | Implement a ChatGPT-like LLM in PyTorch from scratch | End-to-end notebooks for using Weaviate features and integrations |
| Stars | 99k | 939 |
| Forks | 15k | 194 |
| Open issues | 4 | 5 |
| Language | Jupyter Notebook | Jupyter Notebook |
| License | Other | - |
| Last pushed | Jun 2, 2026 | Jun 12, 2026 |
| Categories | Model Training, LLM Frameworks | Vector Databases, Evaluation & Observability, Data & Retrieval, LLM Frameworks |
LLMs-from-scratch
Repository containing code and instructions for developing, pretraining, and finetuning a GPT-like large language model (LLM) using PyTorch.
Jupyter Notebook
recipes
The repository offers comprehensive, ready-to-use datasets and Jupyter Notebook examples to demonstrate various functionalities of Weaviate including vector search, hybrid, and generative AI. It also provides integration materials with external technologies such as LLM frameworks and data platforms.
Jupyter Notebook