ai-agents-huggingface-chat-tem-huggingfacepipeline-does-not-apply-chat-templates--04338762

THEN Use ChatHuggingFace to wrap HuggingFacePipeline, which correctly applies the model's chat template via the tokenizer's apply_chat_template method. For versions of langchain-huggingface ≤0.0.3, manually apply commit 4796b7eb15b4c3a352c950f56a30659f0379f6e2 to include the fix for chat template integration.

from langchain_community.llms.huggingface_pipeline import HuggingFacePipeline
from langchain_community.chat_models.huggingface import ChatHuggingFace

hf_llm = HuggingFacePipeline.from_model_id(
    model_id="TinyLlama/TinyLlama-1.1B-Chat-v1.0",
    task="text-generation",
    pipeline_kwargs={"max_new_tokens": 128}
)
chat_llm = ChatHuggingFace(llm=hf_llm)
# Now use chat_llm with ChatPromptTemplate to get properly formatted prompts.