Unverified50% confidenceBenchmarkExact time
根据 Simon Willison 实测，新分词器导致英文文本 Token 增幅 +42%，西班牙文 +33%，Python 代码 +27%，中文基本无变化
1
Sources
50%
Confidence
Long-term
Relevance
7/6/2026
First Seen
Sources
Is Claude Sonnet 5 Really Cheap? A Cost Breakdown by Task
bilibili为什么叫QQ7/1/2026
Related Claims
UnverifiedLlama 3 的词表扩展使得非英语文本（特别是使用非拉丁字母的语言）的编码效率提升约 30-40%73% similarUnverified研究表明人类阅读字幕时单次注视有效信息量约为7±2个汉字，超过14-16字单行字幕会显著增加认知负荷68% similarUnverifiedLarge language models are billed by the number of tokens, with 1 token equaling approximately 4 English characters or 1–2 Chinese characters.67% similarUnverifiedEnglish data accounts for an overwhelming proportion of mainstream large language model training corpora; taking public datasets such as Common Crawl as an example, English content typically constitutes 40%–60% of the total, while languages such as Vietnamese and Chinese often account for less than 5%.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/113241API
curl https://kongchang.com/api/v1/knowledge/claims/113241MCP
get_claim(id=113241)