Update app.py

app.py

@@ -1,5 +1,6 @@
 from fastapi import FastAPI
 from transformers import AutoModel, AutoTokenizer
+import uvicorn
 
 tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-0.6B")
 model = AutoModel.from_pretrained("Qwen/Qwen3-0.6B")
@@ -18,4 +19,6 @@ def message(req):
         attention_mask=inputs["attention_mask"],
         temperature=0.3
     )
-    return tokenizer.decode(output[0], skip_special_tokens=True)
+    return tokenizer.decode(output[0], skip_special_tokens=True)
+
+uvicorn.run(app, host="0.0.0.0", port=7860)