I Built the Smallest LLM I Could to Understand How They Work
A hands-on walkthrough of a ~108k-parameter character-level GPT in under 200 lines of PyTorch, with the full code included.
#llm
Posts tagged #llm
Write an AI Agent Loop in Under 100 Lines
Build a working coding agent in plain Python: a model, three tools, and a loop that acts, observes, and decides what to do next.
Strata Squeezes a 125-Billion-Parameter AI Onto a Gaming PC. Here's How.
A free, open-source project that runs a huge AI model on a regular gaming graphics card. No cloud, no monthly bill. Here is what it does and how it pulls it off.
Gemini 4 Argon: Google's new flagship is here, and almost nobody can use it yet
It leads on knowledge work, trails on a few coding tests, and is limited to cyber defenders for now. Here's what Google actually shipped.
A 4B Open Model Matches GPT-5.6 Sol on Retrieval at 100× Lower Cost
How Castform and Neon use RL post-training, hybrid search, and existing company data to turn small open models into specialized retrieval agents.
Token Budgeting: The Engineering Skill Nobody Talks About
Token optimization is a context engineering problem, not a prompt-shortening contest — and the two levers most developers never touch will cut your AI bill by 70%
