Unverified50% confidenceBenchmarkExact time
在工具调用测试中，GLM 5.2第一次未能正确调用工具，重试到第三次才成功
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/9/2026
First Seen
Valid until: 10/7/2026
Sources
Sakana Fugu vs GLM 5.2实测：谁能真正挑战顶级AI模型
bilibili赛博门外憨7/7/2026
Related Claims
UnverifiedSonnet 5 拥有 GLM 5.2 所不具备的修订（revision）能力63% similarUnverified原先使用GLM-5.1的用户将被自动路由至GLM-5.2，无需手动切换60% similarUnverified此前Gemma 4:31b在Ollama上存在工具调用失败(failed tool calls)的问题，新版本中基本消失59% similarUnverified在BridgeSpace浏览器代理启动失败的复杂Bug测试中，Composer 2.5和Opus 4.7第一次都未能修复59% similarUnverified此次更新最终移除了模型选择器的键盘契约测试（keyboard contract test）56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/381286API
curl https://kongchang.com/api/v1/knowledge/claims/381286MCP
get_claim(id=381286)