OllamaFrom the builder
Ollama now supports Jev-style decision models
Starting with version 0.35, Ollama supports decision models built on TypeSafe's Jev API, sending text state plus a set of named questions to a local model through a new /v1/systemone endpoint and getting every answer in one request. Three decision models are available at launch: nimble, an open-source 9B model from Bespoke Labs, and Together AI's experimental tev1 in 4B and 0.8B sizes. Ollama says nimble averaged 91ms per decision in a Pac-Man demo running locally on an M5 Max.
Decision models on Ollama are fast, as requests don’t have to travel over a network. Nimble 9B averaged 91ms per decision in the Pac-Man example below when running locally on an M5 Max.
Saved . A shortened excerpt from the official blog; the original publisher retains authorship.