Expired50% confidenceFactExact time
This test uses Cursor International Edition as the unified IDE to conduct a horizontal comparison of four models.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 9/30/2026(expired)
Sources
AI Coding Real-World Test: GPT-5, Gemini 2.5 Pro, Kimi K2, and Grok 4 All Fail at Web Scraping
bilibili图动中国9/18/2025
Related Claims
UnverifiedAISI的评估采用多层次测试框架，参照CVE数据库中的真实案例，同时设计全新挑战场景测试模型泛化能力66% similarUnverifiedAISI's evaluation of GPT-5.5 continued the testing framework previously used for Claude Mythos, ensuring comparability of results across different models.63% similarUnverifiedIntegration testing combines multiple modules that have already passed unit testing and tests them together to verify whether the interactions between modules are functioning correctly.61% similarUnverified先隔离测试集再在训练集内做交叉验证的设计符合TRIPOD等临床预测模型国际报告规范，可规避数据泄漏陷阱60% similarUnverifiedThis evaluation framework includes five benchmarks: ARC-AGI-V2, SWE-Bench, Terminal Bench 2.0, GPQA, and MMLU.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/58532API
curl https://kongchang.com/api/v1/knowledge/claims/58532MCP
get_claim(id=58532)