Add initial implementation of a question answering bot using Gradio and LangChain

.gitignore

@@ -0,0 +1 @@
+venv

app.py

@@ -0,0 +1,38 @@
+import gradio as gr
+from langchain_huggingface.llms import HuggingFacePipeline
+from langchain.prompts import PromptTemplate
+
+# Initialize the model pipeline
+llm = HuggingFacePipeline.from_model_id(
+    model_id="ibm-granite/granite-3.2-2b-instruct",
+    task="text-generation",
+    pipeline_kwargs={
+        "max_new_tokens": 512,
+    },
+)
+
+prompt = PromptTemplate(
+    input_variables=["question"],
+    template="Answer the following question: {question}",
+)
+
+# Create a chain with the model pipeline and prompt template
+chain = prompt | llm
+
+def ask_question(question: str) -> str:
+    # Invoke your chain and return the generated answer.
+    response = chain.invoke({"question": question})
+    return response
+
+# Create a Gradio interface
+iface = gr.Interface(
+    fn=ask_question,
+    inputs="text",
+    outputs="text",
+    title="Question Answering Bot",
+    description="Ask any question and get an answer from the model."
+)
+
+# Launch the Gradio interface
+if __name__ == "__main__":
+    iface.launch()

requirements.txt

@@ -0,0 +1,3 @@
+gradio
+langchain
+langchain_huggingface