Unverified50% confidenceBenchmarkExact time
Decision-1 has a calibration score of 92.2 (out of 100), is approximately 2.5 times faster than the second-ranked decision model, and approximately 35 times faster than GPT-6 Sol.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/11/2026
First Seen
Valid until: 1/9/2027
Sources
Related Entities
Related Claims
UnverifiedTaking GPT-MoE as an example, if the total number of experts is 64 and each token only activates Top-2, the actual computation can be reduced by approximately 97% compared to a Dense model of equivalent scale70% similarUnverifiedWhen using the multi-agent verification process, MARCH's accuracy was suppressed to 4.4%, whereas when the same model was directly asked which solution was better, accuracy reached 43.7%.69% similarUnverifiedThe official core metrics for Decision Models are: an average of 90 milliseconds per decision, approximately 76% accuracy on public benchmarks, and free local execution.68% similarUnverifiedCompared to standard LoRA, counterfactual mDPO improved decision accuracy from 93.33% to 96.67%.66% similarUnverifiedIn an editing benchmark, the same GPT model's score dropped from 99% to 70% depending on the framework used.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/998000API
curl https://kongchang.com/api/v1/knowledge/claims/998000MCP
get_claim(id=998000)