Logan Olson
Logan Olson
Does either option work with nn.MLP or would I need to do more surgery? It's unclear to me how I would use tree_at to access each linear layer but maybe...
It seems to just work out of the box if you put a streamer in your pipeline: ``` streamer = TextStreamer(tokenizer) pipe = pipeline(model=model, tokenizer=tokenizer, streamer=streamer} llm = HuggingFacePipeline(pipeline=pipe) ```
What needs to happen to get this grammar pull request approved and merged? I'd love to start using grammars with vLLM.