// HACKER NEWS — CYBERSECURITY
Ollaya – Ollama for open-source, Jev-style decision models
Ask typed questions about any text or JSON and get calibrated answers in milliseconds. Private, open source, on your own hardware.
A decision model answers in a single forward pass, with no token-by-token generation. On your own GPU, a five-question request to Laya takes about 10 ms, end to end through the HTTP API.
Ollaya: median of a five-question request through the HTTP API on an NVIDIA RTX 4090 (laya in fp16, the others in fp32). Jev: median request latency of the hosted API in third-party benchmarks (AbdelStark/jev-benchmarks, nibzard/decision-model-benchmark), which includes the network. Setups differ, so read it as an order-of-magnitude comparison.
Ollaya serves /v1/systemone and /v1/models with TypeSafe's request and response shapes. The official TypeSafe Python SDK 0.7.1 works unchanged against a local server.
Start with Laya from Convai Innovations: an English model, a 100+ language model, a model fine-tuned for typed decisions, and a router that picks for you.
More open decision models are planned: von, GGUF LLM-based decision models via llama.cpp.
Tickets, emails and user messages are often the most sensitive data you have. With Ollaya they are scored where they already live.
Runs on your machine with ONNX Runtime, on the CPU or an NVIDIA GPU. The server listens on 127.0.0.1 by default.
Weights come from their authors’ Hugging Face repositories, pinned to a commit and checked against sha256. Ollaya never re-hosts them, and the runtime is Apache-2.0.
Run as many decisions as your hardware can handle. No metering and no API bill.