Collection of Open Source Projects Related to GPT，GPT相关开源项目合集🚀、精选🔥🔥
MiniGPT-16MB: The Ultra-Light Storyteller, driven by a high-performance Language Model (LLM) designed to run entirely in the browser. At under 16MB, this model is optimized for low-end hardware, mobile devices, and offline use.
Independent MiniGPT research lab: a from-scratch Transformer, training and inference components, and reproducible toy-scale experiments in grokking and interpretability.
From-Scratch High Valyrian MiniGPT — Build and train a GPT-like transformer model fully from scratch, including custom tokenizer, dataloader, transformer, training, and inference! Learn the internals of LLMs layer-by-layer.
A complete implementation of a GPT-style decoder-only Transformer language model built from scratch in PyTorch, including BPE tokenization, dataset preprocessing, training pipeline, mixed-precision training, checkpointing, and text generation.
Course work from UCLA's ECE239 - Special Topics in Signals and Systems
An interactive showcase for a custom-built Mini-GPT model, featuring architecture details, training visualizations, and a text generation playground. · Built with Manus
Train, run, and configure your own local AI from scratch! This includes a full frontend styled like the ChatGPT website, automatic hyperparameters finetuned for your specific device and training time, and full tokenization handling.
A mini version of GPT implemented on shakespear using BPE
A Mini GPT-like Language Model built from scratch using PyTorch, implementing tokenization, embeddings, self-attention, transformer layers, training, and text generation.
From-scratch mini-GPT trained on live Bluesky data via Jetstream [30B]
A hands-on suite for exploring and fine-tuning foundation models (Transformers, BERT, GPT-2, BART) and end-to-end RAG pipelines with attention visualizations, semantic search (ChromaDB/Weaviate), LangChain workflow demos.
To associate your repository with the minigpt topic, visit your repo's landing page and select "manage topics."