Atualiza app.py para usar modelo Qwen-4B-Instruct

app.py

@@ -1,19 +1,44 @@
 import os
+from fastapi import FastAPI
+from pydantic import BaseModel
 from openai import OpenAI
+import langdetect  # pip install langdetect
+
+app = FastAPI()
 
 client = OpenAI(
     base_url="https://router.huggingface.co/v1",
     api_key=os.environ["HF_TOKEN"],
 )
 
-completion = client.chat.completions.create(
-    model="meta-llama/Llama-3.1-8B-Instruct:fireworks-ai",
-    messages=[
-        {
-            "role": "user",
-            "content": "What is the capital of France?"
-        }
-    ],
-)
+class Query(BaseModel):
+    question: str
+
+@app.get("/")
+async def root():
+    return {"message": "API está rodando!"}
+
+@app.post("/ask")
+async def ask_model(query: Query):
+    try:
+        idioma = langdetect.detect(query.question)
+    except:
+        idioma = "pt"
+
+    if idioma == "pt":
+        system_prompt = "Responda sempre em português, de forma curta e direta, sem explicações longas."
+    elif idioma == "en":
+        system_prompt = "Answer always in English, briefly and directly, no long explanations."
+    else:
+        system_prompt = "Responda no mesmo idioma da pergunta, de forma direta e curta."
 
-print(completion.choices[0].message)
+    completion = client.chat.completions.create(
+        model="meta-llama/Llama-3.1-8B-Instruct:fireworks-ai",
+        messages=[
+            {"role": "system", "content": system_prompt},
+            {"role": "user", "content": query.question}
+        ],
+        max_tokens=150
+    )
+    answer = completion.choices[0].message.content
+    return {"answer": answer}