Laya (Fine-Tuned on Typed-Decisions + mybillbook KB-routing)

Laya fine-tuned on the LocalLLaMA/typed-decisions benchmark (1,200 cases) plus a tenant-specific KB-routing dataset built from real vocabulary collisions in mybillbook'''s support knowledge bases (printer vs billing, gst vs reports, subscription vs login, etc. - see github FlutterBot repo laya/datasets/README.md for the full analysis).

Directly fixes the printer-vs-billing misclassification bug: "my application is failing to print bill" now classifies as printer (0.70+ confidence) instead of confidently misrouting to billing as the base/generic-fine-tuned model did.

Downloads last month: -; Downloads are not tracked for this model. How to track

Safetensors

Model size

0.4B params

Tensor type

F16

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support