Challenging Tokenization: An LLM Architecture Experiment Without a Tokenizer

For years, most Large Language Models (LLMs) have started from the same assumption: Language is first broken into tokens, and computation happens on those tokens.

Whether it's BPE, SentencePiece, or WordPiece, the tokenizer remains one of the most fundamental components of the entire architecture.

But while working on a new experimental framework called NINMENI, we decided to ask a different question: What if tokenization isn't the only possible foundation?

Instead of treating tokens as the primitive unit, NINMENI explores a different assumption:

1 Character = 1 ID