William Liu
A developer and researcher who, in the tiktoken repository, compared the tokenization results of multiple GPT models through fixture experiments.
Timeline (last 90 days)
William Liu compared the tokenization results of multiple GPT models using a set of fixtures, with the experiment covering seven GPT models, including GPT-5.5, three variants of GPT-5.6 (Sol/Terra/Luna), and three variants of GPT-6 (Astra/Sol/Luna).
In William Liu's experiment, across all 31 test samples, the token counts reported by the seven GPT models were completely identical, all at 44,794.
All Facts (2)
In William Liu's experiment, across all 31 test samples, the token counts reported by the seven GPT models were completely identical, all at 44,794.
50%UnverifiedWilliam Liu compared the tokenization results of multiple GPT models using a set of fixtures, with the experiment covering seven GPT models, including GPT-5.5, three variants of GPT-5.6 (Sol/Terra/Luna), and three variants of GPT-6 (Astra/Sol/Luna).
50%