Expired50% confidenceBenchmarkExact time
In tool-calling tests, GLM 5.2 failed to call the tool correctly on the first attempt and did not succeed until the third retry
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/9/2026
First Seen
Valid until: 10/7/2026(expired)
Sources
Sakana Fugu vs GLM 5.2 Hands-On Test: Who Can Truly Challenge Top-Tier AI Models?
bilibili赛博门外憨7/7/2026
Related Claims
UnverifiedSonnet 5 has revision capability that GLM 5.2 lacks.63% similarUnverifiedUsers previously on GLM-5.1 will be automatically routed to GLM-5.2 without needing to switch manually.60% similarUnverifiedPreviously, Gemma 4:31b had issues with failed tool calls on Ollama, which have largely disappeared in the new version.59% similarUnverifiedIn a complex bug test involving a BridgeSpace browser agent startup failure, both Composer 2.5 and Opus 4.7 failed to fix it on the first attempt59% similarUnverifiedThis update ultimately removed the keyboard contract test for the model selector.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)