Bring Your Own Model #

ollama #

Running models locally with ollama is supported. Using ollama as a local model manager and runner is supported and after downloading and installing ollama EndpointElf can use the local models.

Run a model locally,

ollama run qwen2.5

ollama run codellama

Hint: Take advantage of performance improvements that ollama supports and their FAQ.

Running EndpointElf #

With your model running include it in EndpointElf’s configuration. If you want to only extract routes with AI add -ai=false to the command.

endpointelf -ai.engine ollama -output api.yaml update ./code/

Include -ai.url=http://127.0.0.1:11434 to override the where EndpointElf communicates with an AI agent.