llm-semantic-router

/

Decision-1.0-Route-0.6B

Like 0

Follow

vLLM Semantic Router 204

Adds the model's own inference code so AutoModel.from_pretrained(repo, trust_remote_code=True) and pipeline("decision", ...) run it locally; system_one(state=..., questions=...) takes and returns System One request and response bodies. config.json keeps every Decision key and gains model_type, architectures, auto_map and custom_pipelines; the card gains a short "Use with 🤗 Transformers" section. Weights, tokenizer and every other file are unchanged.