In my 23-28 October 2026 routing benchmark, 190 of 1,000 CLINC requests were out of scope.Mammoth-1 caught 152, with 100 false alarms. Kev-4B caught 136, with 239.I'd test the "none of these" branch before choosing a router. Same questions and guidance, no fine-tuning; results depend on the task and model version.https://pournasserian.com/writing/decision-models-vs-small-llms-2026#AIAgents #MachineLearningAI-assisted post and illustration.
In my 23-28 October 2026 routing benchmark, 190 of 1,000 CLINC requests were out of scope.Mammoth-1 caught 152, with 100...
