Expired60% confidenceFactExact time
English content typically accounts for 60%–80% of the training data in mainstream large language models.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Related Claims
UnverifiedEnglish 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%.84% similarUnverifiedEnglish content in mainstream large language model training corpora typically exceeds 50%, leading to significant performance gaps for non-English languages.82% similarUnverifiedEnglish-language data accounts for the highest proportion in most large language models, typically exceeding 50%, so writing prompts in English may yield better results on certain complex tasks.76% similarUnverifiedMeta Llama 2的2万亿token训练数据中英语占比约89.7%，欧洲语言合计约8%，其余所有语言共享约2%76% similarUnverifiedSome large language models have already achieved over 80% accuracy on competition-level problem sets such as MATH and AIME.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/43711API
curl https://kongchang.com/api/v1/knowledge/claims/43711MCP
get_claim(id=43711)