Expired60% confidenceFactExact time
GPT-4o introduced a new tokenizer optimized for multilingual use, improving Chinese token efficiency by approximately 3 times, with roughly 1–1.5 tokens now corresponding to one Chinese character.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Is ChatGPT Pro Account Sharing Worth It? 3 Major Risks and 4 Safer Alternatives
bilibili云不加冰6/26/2026
Related Claims
UnverifiedThe improvement in GPT 5.5's Chinese comprehension capabilities is attributed to a significantly increased proportion of Chinese-language corpus in the training data and the introduction of more Chinese annotators' preference data during the RLHF stage.75% similarUnverifiedGPT-4的tiktoken分词器中，一个英文单词平均对应1.3个token，每个汉字约占2-3个token74% similarUnverifiedGPT-4o is more balanced in multilingual mixed scenarios74% similarVerifiedGPT-4o supports a context window of 128K Tokens, capable of processing approximately 100,000 Chinese characters.73% similarUnverifiedGPT-4所使用的cl100k_base词表包含约10万个Token单元，对中文单字的覆盖率相对较低72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/61641API
curl https://kongchang.com/api/v1/knowledge/claims/61641MCP
get_claim(id=61641)