[KongchangAI]
Unverified50% confidenceBenchmarkExact time

Native decision models achieve 62.50% accuracy on the TandemBench benchmark, but are close to random on optimality verification.

1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/5/2026
First Seen
Valid until: 2/3/2027

Sources

Related Entities

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/1137706
API
curl https://kongchang.com/api/v1/knowledge/claims/1137706
MCP
get_claim(id=1137706)