Expired50% confidenceFactExact time
In standardized tests, GPT-4's performance has surpassed that of more than 90% of human test-takers.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Andrew Ng's AI Prompting Guide: Four Key Gaps Between Beginners and Experts
bilibili吴恩达_AndrewNg6/4/2026
Related Claims
UnverifiedEnglish accounts for more than 90% of GPT-4's training data.85% similarUnverifiedGPT-4在美国执照医师考试USMLE中通过率超过90%83% similarVerifiedModels such as GPT-4, Claude 3.5, and DeepSeek have achieved pass rates exceeding 80% on the HumanEval programming benchmark.79% similarUnverified研究表明GPT-4作为评估者与人类专家评分的一致性超过80%，显著高于BLEU、ROUGE等基于词汇匹配的自动化指标78% similarUnverifiedGPT-5.4 achieved a score of 75% on the OS World Verified benchmark, surpassing the human baseline of 72.4% for the first time.78% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/52669API
curl https://kongchang.com/api/v1/knowledge/claims/52669MCP
get_claim(id=52669)