All projects

Related projects

  • groot

    🌳 Groot 1 — open typed-decision model. Calibrated choice/score/yes-no answers over 64k-token states, System One API compatible, with a pure Bend engine. Beats Jev on zero-shot Folha news (87.8% vs 85.8%) and typed decisions (75.3% vs 73.7%).

    Shared topics: machine-learning

  • laya-mlx

    Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.

    Shared topics: machine-learning

  • jev-torch

    PyTorch implementation for training JEV-style calibrated decision models

    Shared topics: machine-learning