Ollaya – Ollama for open-source, Jev-style decision models

Ollaya runs open-source decision models locally on NVIDIA GPUs or CPUs using ONNX Runtime for private, millisecond-latency data processing. It maintains drop-in compatibility with the TypeSafe API, allowing developers to execute typed queries without token-by-token generation. Users can choose from various models like Laya, Decider, and Qwen3guard to perform calibrated, zero-shot classification tasks.