Unverified50% confidenceBenchmarkExact time
At the 64k vocabulary setting, 88.6% of words were retained as a single token.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/17/2026
First Seen
Valid until: 2/15/2027
Sources
Related Entities
Related Claims
UnverifiedAt the 64k vocabulary setting, the new tokenizer averaged 1.147 tokens per word, beating the best existing tokenizer's 1.208 and XLM-RoBERTa's 1.512.74% similarUnverifiedIn a warm English encoding with bert-base-uncased, Normalization consumed 88% of the encoding time, while WordPiece accounted for only 8%.69% similarUnverifiedWordCraft is currently in alpha; the author estimates that by command count it covers 88% of Word's ribbon features, but it is only about 60% of the way to truly replacing Word, and for everyday basic needs that figure is 63%.68% similarUnverifiedTinyLlama-1.1B's tokenizer produces an average of 1.34 subword tokens per word when processing Latin script, and an average of 22.60 tokens per word when processing Kannada.64% similarUnverifiedEnglish perplexity is 6.6088, compared with the baseline of 6.3638.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1178559API
curl https://kongchang.com/api/v1/knowledge/claims/1178559MCP
get_claim(id=1178559)