Upload app/main.py with huggingface_hub

app/main.py

@@ -0,0 +1,27 @@
+
+from fastapi import FastAPI
+from pydantic import BaseModel
+from app.model import model, tokenizer
+import torch
+
+app = FastAPI()
+
+class Prompt(BaseModel):
+    text: str
+
+@app.post("/generate")
+def generate(prompt: Prompt):
+    full_prompt = f"<s>[INST] {prompt.text.strip()} [/INST]"
+    inputs = tokenizer(full_prompt, return_tensors="pt").to("cuda" if torch.cuda.is_available() else "cpu")
+
+    with torch.no_grad():
+        outputs = model.generate(
+            **inputs,
+            max_new_tokens=256,
+            temperature=0.7,
+            top_p=0.9,
+            do_sample=True
+        )
+
+    result = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return {"response": result.split("[/INST]")[-1].strip()}