|
from fastapi import FastAPI, HTTPException |
|
from pydantic import BaseModel |
|
from huggingface_hub import InferenceClient |
|
import os |
|
from google import genai |
|
|
|
app = FastAPI() |
|
|
|
|
|
hf_token = os.environ.get("HF_TOKEN") |
|
google_api_key = os.environ.get("GOOGLE_API_KEY") |
|
|
|
class ChatRequest(BaseModel): |
|
message: str |
|
system_message: str = """You are Dan Infalt, a public land deer hunting expert specializing in targeting mature bucks in pressured areas. |
|
You focus on buck bedding, terrain reading, and aggressive yet calculated mobile tactics. Your blue-collar, no-nonsense approach |
|
emphasizes deep scouting, strategic access, and minimalist setups. Through The Hunting Beast, you teach hunters how to kill big bucks |
|
using terrain, wind, and thermals. You speak from firsthand experience, keeping your advice practical and to the point. Provide detailed |
|
yet concise responses, with a maximum of 150 words""" |
|
max_tokens: int = 512 |
|
temperature: float = 0.7 |
|
top_p: float = 0.95 |
|
model_choice: str = "HF" |
|
|
|
class ChatResponse(BaseModel): |
|
response: str |
|
|
|
prompt_template = f"""""" |
|
|
|
@app.post("/chat", response_model=ChatResponse) |
|
async def chat(request: ChatRequest): |
|
try: |
|
if model_choice == "HF": |
|
if hf_token: |
|
client = InferenceClient("meta-llama/Llama-3.2-3B-Instruct", token=hf_token) |
|
else: |
|
raise ValueError("HF_TOKEN environment variable not set. Please add it as a secret in your Hugging Face Space.") |
|
|
|
messages = [ |
|
{"role": "system", "content": request.system_message}, |
|
{"role": "user", "content": request.message}, |
|
] |
|
|
|
response = client.chat_completion( |
|
messages=messages, |
|
max_tokens=request.max_tokens, |
|
temperature=request.temperature, |
|
top_p=request.top_p, |
|
) |
|
|
|
return {"response": response.choices[0].message.content} |
|
|
|
if model_choice == "google": |
|
genai.configure(api_key=google_api_key) |
|
model = genai.GenerativeModel("gemini-2.0-flash") |
|
|
|
messages = [ |
|
{"role": "system", "parts": [request.system_message]}, |
|
{"role": "user", "parts": [request.message]}, |
|
] |
|
|
|
response = model.generate_content(messages) |
|
|
|
if response and hasattr(response, 'text'): |
|
return response.text |
|
else: |
|
return "No response text received from the model." |
|
|
|
except Exception as e: |
|
raise HTTPException(status_code=500, detail=str(e)) |
