Unverified50% confidenceBenchmarkExact time
Mythos scored 64.7% on Humanity's Last Exam, far exceeding any currently publicly available model.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/8/2026
First Seen
Valid until: 1/6/2027
Sources
Related Entities
Related Claims
UnverifiedThe 2023 TruthfulQA benchmark showed that even the strongest models at the time did not exceed 60% accuracy on questions requiring precise factual judgment.62% similarUnverifiedIthaca模型缺失文字预测准确率约62%，人类专家独立完成时约为25%，地理来源推断约71%，年代估算误差中位数约30年61% similarUnverified在HumanEval等专业代码基准测试中，顶尖模型的表现已接近甚至超越人类平均水平58% similarUnverifiedA single "trusted source" prompt caused 7 out of 8 tested models to reverse between 45% and 88% of their correct answers.58% similarUnverifiedRoboDojo具身智能测评显示，最强模型得分仅12.8分（人类基准为100分），仿真成功率不足9%，真机成功率12.8%58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/989591API
curl https://kongchang.com/api/v1/knowledge/claims/989591MCP
get_claim(id=989591)