Expired50% confidenceBenchmarkExact time
GLM-5.2 answered incorrectly on first attempt in a character-counting test for 'how many S's are in occasion,' but corrected its answer to 2 after enabling DeepThink deep reasoning.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/9/2026
First Seen
Valid until: 10/7/2026(expired)
Sources
The Complete Guide to GLM-5.2: Hands-On Review of an Open-Source Long-Context Model
bilibili英文白斑马7/4/2026
Related Claims
UnverifiedGLM-4.6's tool calls failed in reasoning mode, and the evaluation team had to disable reasoning mode to get it running properly.67% similarUnverifiedThe GLM series uses an Autoregressive Blank Infilling training objective, which randomly masks text spans and requires the model to autoregressively restore them60% similarUnverifiedSW1E1's official comparison report does not include cross-comparisons with mainstream reasoning models such as DeepSeek R1 and Qwen 3.57% similarUnverifiedChatGPT注意到原始矩M_n与泰勒/Jensen系数之间存在阶乘归一化的不一致，该偏差严重到使论文某结论似乎与已知定理相矛盾56% similarUnverifiedRAG原始论文在多个知识密集型基准测试上超越了当时的纯参数化模型55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/331744API
curl https://kongchang.com/api/v1/knowledge/claims/331744MCP
get_claim(id=331744)