Verified70% confidenceFactExact time
Claude在编程基准测试（如HumanEval、SWE-bench等）中持续名列前茅
4
Sources
70%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Cursor AI编程工具保姆级教程：零基础实战指南
bilibili子涵Python6/25/2026
Related Claims
VerifiedClaude模型在SWE-bench等编程能力基准测试中持续领先78% similarVerifiedClaude系列在SWE-bench（软件工程基准测试）等权威评测中持续位居前列75% similarUnverifiedAnthropic在代码基准测试（如HumanEval、SWE-bench）上持续领先72% similarUnverifiedClaude 系列模型在 SWE-bench 等编码基准测试中长期位居前列70% similarVerifiedClaude Code依托的Sonnet模型在HumanEval、SWE-bench等主流编程基准测试中持续领先70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/41223API
curl https://kongchang.com/api/v1/knowledge/claims/41223MCP
get_claim(id=41223)