from huggingface_hub import InferenceClient
from llama_index.core.tools import FunctionTool

# --- Functions --- #

def query_image(query: str, image_url: str) -> str:
    """Ask anything about an image using a Vision Language Model

    Args:
        query (str): the query about the image, e.g. how many persons are on the image?
        image_url (str): the URL to the image
    """

    client = InferenceClient(provider="nebius")
    try:
        completion = client.chat.completions.create(
            # model="google/gemma-3-27b-it",
            model="Qwen/Qwen2.5-VL-72B-Instruct",
            messages=[
                {
                    "role": "user",
                    "content": [
                        {
                            "type": "text",
                            "text": query
                        },
                        {
                            "type": "image_url",
                            "image_url": {
                                "url": image_url
                            }
                        }
                    ]
                }
            ],
            max_tokens=512,
        )
        return completion.choices[0].message

    except Exception as e:
        return f"query_image failed: {e}"


def automatic_speech_recognition(file_url: str) -> str:
    """Transcribe an audio file to text

    Args:
        file_url (str): the URL to the audio file
    """
    client = InferenceClient(provider="fal-ai")
    try:
        return client.automatic_speech_recognition(file_url, model="openai/whisper-large-v3")
    except Exception as e:
        return f"automatic_speech_recognition failed: {e}"



### --- Tool instance ---

query_image_tool = FunctionTool.from_defaults(query_image)
automatic_speech_recognition_tool = FunctionTool.from_defaults(automatic_speech_recognition)