Unverified50% confidenceBenchmarkExact time

GLM-5.2官方称在部分真实测试中超越Opus，并在多项编程基准上达到开源SOTA

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

7/8/2026

First Seen

Valid until: 10/6/2026

Sources

AI早报：600亿收购Cursor、GLM-5.2开源、DeepSeek首轮融资超500亿

bilibili橘鸦Juya6/17/2026

Related Claims

Unverified截至发稿时，GLM-5.2尚未公布任何官方跑分数据，网上流传的高分均来自GLM-5或GLM-5.172% similar Unverified一位Reddit用户在使用某AI工具时遇到提示信息'Prepared using Best because GLM-5.2 was inapplicable or unavailable'68% similar Unverified评测基准面临数据污染问题，训练数据可能已包含测试题目导致分数虚高，GPT-4技术报告中OpenAI首次公开讨论了这一问题60% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/212920

API

curl https://kongchang.com/api/v1/knowledge/claims/212920

MCP

get_claim(id=212920)