Verified65% confidenceFactTime unknown
MBPP(Mostly Basic Python Problems)由Google发布,包含约1,000个入门级编程问题
3
Sources
65%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Crafta-Bench:Cursor后台Agent基准测试工具深度解析
githubnwinter
Related Entities
Related Claims
UnverifiedMBPP 全称为 Mostly Basic Python Problems,是一种考察模型独立封闭问题表现的代码基准79% similarUnverifiedMBPP代码生成基准测试包含约500道面向初级程序员的Python编程题74% similarUnverifiedMBPP包含974个基础Python任务,评估指标通常为pass@k61% similarUnverified微软Phi-1模型(1.3B参数)在Python代码生成任务上超越了参数量大数十倍的竞品59% similarUnverified100万token上下文大约相当于2-3个Flask规模的完整项目,或十几个常见Python包的全部代码55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/31099API
curl https://kongchang.com/api/v1/knowledge/claims/31099MCP
get_claim(id=31099)