@owlmeans/llm

LLM inference runtime for OwlMeans applications: a resilient four-method model over any LangChain chat model, a provider-plugin layer, a model factory service, and an execution abstraction that resolves models from an inheritable policy.