Local LLM
This is a PowerToys Run plugin that lets a Windows user query a local Ollama model directly from the PowerToys Run launcher, choosing which pulled model to ask on the fly.
The plugin requires Ollama to already be installed and running locally, along with at least one pulled model such as Llama 3.1, and sends a query typed into PowerToys Run to Ollama's local API. A query can either use whichever model is configured as the default in the plugin's settings, or explicitly pick a different pulled model by name as part of the query itself.
Response time depends on the user's own GPU, since inference runs entirely on their machine through the locally running Ollama server rather than a cloud API.
It is written in C# and does not yet have an ARM64 Windows build, since Ollama itself has none.