Unverified50% confidenceBenchmarkExact time
Qwen and Flash-Lite are the models most prone to wild guessing on cold cases, with wild-guess rates of 59% and 45% respectively.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/11/2026
First Seen
Valid until: 10/25/2026
Sources
Related Entities
Related Claims
UnverifiedAdding a rule book to the prompt drove the wild-guess rate to zero for seven of the eight models, and reduced Qwen from 59% to 7%.58% similarUnverifiedIn a blind evaluation battle designed by the author consisting of 40 questions and three groups of random seeds, SME ranked first with a score of 96.7%, with competitors including MEM0, GRAFFITY, and five other mainstream memory solutions.57% similarUnverifiedQwen3.8 Flash Next has a sparsity rate of approximately 95%, using less than 5% of its parameters per inference.55% similarVerifiedThe per-token inference cost of Flash models is typically one order of magnitude lower than that of Pro-level models.55% similarUnverifiedMuse Glimmer在Artificial Analysis第二版测试中幻觉率高达82%，知识工作和幻觉是其最大弱点54% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/999560API
curl https://kongchang.com/api/v1/knowledge/claims/999560MCP
get_claim(id=999560)