⬢github Rust · 702 ★ +12 since we first saw it · pushed 1 h ago · Apache-2.0
ollaya-dev/ollaya
Run open decision models locally: pull and serve Laya, decider, NLI and GLiClass behind a TypeSafe-compatible API. Ollama for decision models.
Ollaya is a Rust tool that runs open decision models locally, like Ollama does for LLMs. Decision models take a state plus typed questions and return calibrated probabilities in milliseconds without generating text. Ollaya pulls models by name, serves them via a local daemon, speaks TypeSafe's wire format, offers an MCP server for agents, and supports Modelfiles, ONNX/CUDA, and GGUF via llama.cpp.
Why now: It appeared on GitHub's new, most-starred repos list shortly after release, gaining fast traction (690 stars) as a novel 'Ollama for decision models' with local, TypeSafe-compatible inference.
Who it is for: Developers building classification, routing, or triage features who want fast, private, local decision inference without LLM text generation.
calibrationclassificationdecision-modelsgliclassjevlayallm-routinglocal-inferencenliollama
Stars over our 8 snapshots: 690 to 702, since 1 h ago.
Where people talked about it
- ⬢github new repos, most starred 3 min ago
API: https://socialmediatrends-api.osmike.com/v1/repos/ollaya-dev/ollaya