RustGPT: A pure-Rust transformer LLM built from scratch

I replaced Animal Crossing's dialogue with a live LLM by hacking GameCube memory

Llama-Factory: Unified, Efficient Fine-Tuning for 100 Open LLMs

DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning

Use the Accept Header to Serve Markdown Instead of HTML to LLMs

LLMs are still surprisingly bad at some simple tasks

LLM Inference Economics from First Principles

You're not using LLMs enough

LLM Visualization

Tiny LLM - LLM Serving in a Week