LocalLLaMA/typed-decisions
A structured, probabilistic classification benchmark for evaluating a model's multi-dimensional classification and calibration-aware decision-making in enterprise workflow scenarios. It includes subsets such as agent_trace_observability, customer_service, invoice_processing, and security_incidents.
Timeline (last 90 days)
Darwin-27B-ZTC-v5 achieved an accuracy score of 0.81 on the LocalLLaMA/typed-decisions benchmark on Hugging Face.
Darwin-27B-ZTC-v5 topped the LocalLLaMA/typed-decisions leaderboard with a score of 0.81, the highest score ever recorded on that leaderboard.
The LocalLLaMA/typed-decisions dataset has been made public on Hugging Face, can be loaded via the Datasets library, and supports interactive exploration in Data Studio and querying with SQL Console.
All Facts (3)
Darwin-27B-ZTC-v5 achieved an accuracy score of 0.81 on the LocalLLaMA/typed-decisions benchmark on Hugging Face.
50%UnverifiedDarwin-27B-ZTC-v5 topped the LocalLLaMA/typed-decisions leaderboard with a score of 0.81, the highest score ever recorded on that leaderboard.
50%UnverifiedThe LocalLLaMA/typed-decisions dataset has been made public on Hugging Face, can be loaded via the Datasets library, and supports interactive exploration in Data Studio and querying with SQL Console.
50%