Enables support for vLLM. To use, you must specify themodel field in thetask parameter of thepgml.transform functionand you must add"backend": "vllm" in thetask parameters. For example,

SELECT*FROMpgml.transform(    task=>'{"model":"tiiuae/falcon-7b","backend":"vllm"}'::JSONB,    inputs=> Array['hello']);

A list of supported models for vLLM can be foundhere.

Only one vLLM model can be loaded per client connection process due to alimitation in vLLM. The first call topgml.transform with a given model will load the model ("cold start"), but subsequent calls will use the cached model. If you change the specified model in the same client connection, the cached model will be replaced with the new one.