Unverified50% confidenceFactExact time
The Wilson score confidence lower bound formula used by Folkbench is: Score = (p̂ + z²/(2n) − z·√(p̂(1−p̂)/n + z²/(4n²))) / (1 + z²/n), where p̂ is the observed success rate, n is the sample size, and z is the normal quantile corresponding to 95% confidence.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe Z-score is calculated as (observed value − mean) / standard deviation, and an absolute value exceeding 2 or 3 is typically regarded as anomalous.65% similarUnverifiedLoading a model with N parameters at FP16 precision requires approximately 2N bytes of video memory.61% similarUnverifiedAt the same compute cost, an even N is better than its adjacent odd N: at p=0.10, the error rate is 0.0280 for N=3, 0.00370 for N=4, and 0.00856 for N=5.58% similarUnverifiedMuse Glimmer scores 75.5 on the MCP Atlas tool-calling benchmark, leading Qwen 3.6; scores 74.6 on agentic search tasks; and scores 94.7 on the AIME mathematical reasoning benchmark.58% similarUnverifiedQwen2.5-3B-Instruct achieved an overall score of 0.90 on EvidenceFlip, with an action accuracy of 0.90, conflict detection of 1.00, and pressure resistance of 1.00.56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1031001API
curl https://kongchang.com/api/v1/knowledge/claims/1031001MCP
get_claim(id=1031001)