Unverified50% confidenceBenchmarkExact time
MBPP 收录了 500 道众包编程题,GSM8K 包含 8500 道小学数学应用题(需 2-8 个推理步骤)
1
Sources
50%
Confidence
Long-term
Relevance
7/15/2026
First Seen
Sources
Related Claims
UnverifiedHumanEval包含164道编程题,MBPP包含约500道编程题66% similarVerifiedMMLU contains approximately 15,000 multiple-choice questions across 57 subject areas; HumanEval, designed by OpenAI, contains 164 Python programming problems; MATH contains 12,500 competition math problems; and GPQA covers graduate-level science questions.62% similarUnverifiedKimi K2.6 supports releasing 300 parallel sub-agents at once and can generate more than 100 files from a single prompt.59% similarUnverified100万Token意味着模型理论上可一次性处理约50-75万汉字53% similarUnverifiedKimi K2.5's training data contains approximately 15 trillion tokens of mixed visual and text data.52% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/518044API
curl https://kongchang.com/api/v1/knowledge/claims/518044MCP
get_claim(id=518044)