Is it allowed to build Ollama docker to run different model?

Meta Comprehensive RAG Benchmark: KDD Cup 2-9d1937 Meta KDD Cup 24 - CRAG - Knowledge Graph and Web R
#1

I want to build an Ollama docker. And use my model to request Ollama API to get related answer? Is it allowed?

#2

@lu_kun : Yes you are allowed to do that. We nevertheless include a vLLM example with batched offline inference, in the new baseline, which might be simpler to build on top of.