#from-scratch

Faithful from-scratch PyTorch reproduction of OpenAI's GPT-OSS architecture (sliding/full attention alternation, learned attention sinks, YaRN 128K, top-2-of-8 MoE), scaled to Chinchilla-optimal 502M total / 247M active training on a single A100 80GB

Monk LLM Gateway: one API in front of every model. An LLM gateway and router built from scratch (proxy, routing, retry/fallback, caching, budgets), with an interactive course.

A 1.32B-parameter Mixture-of-Experts LLM trained from scratch in PyTorch - custom tokenizer, MLA attention, 18B-token pretraining, SFT and DPO. No pretrained weights, no HuggingFace Trainer.

Mac-local LLM-from-scratch lab for building and evaluating a small decoder-only language model on Apple Silicon.

An LLM inference engine written from scratch in Python and C++/Metal — tokenizer, forward pass, KV cache, and INT4 quantization implemented by hand. No PyTorch, no llama.cpp.

JimmyLabs — a GPT-style language model built from scratch and fully documented on a MacBook Air M1 (8GB, Apple Silicon). Documentation-first educational research platform.