Update utils/inference.py

utils/inference.py

@@ -12,13 +12,15 @@ headers = {
 
 def call_model(prompt: str) -> str:
     response = requests.post(
-        f"{API_URL}/v1/completions",
+        f"{API_URL}/generate",  # <-- use /generate for HF endpoints
         headers=headers,
         json={
-            "prompt": prompt,
-            "max_new_tokens": 2048,
-            "temperature": 0.3,
-            "do_sample": False
+            "inputs": prompt,
+            "parameters": {
+                "max_new_tokens": 2048,
+                "temperature": 0.3,
+                "do_sample": False
+            }
         }
     )
 
@@ -26,4 +28,13 @@ def call_model(prompt: str) -> str:
         raise RuntimeError(f"Inference error: {response.status_code} - {response.text}")
 
     result = response.json()
-    return result.get("generated_text", "⚠️ No output generated.")
+    
+    # Handle variations in response format
+    if isinstance(result, dict) and "generated_text" in result:
+        return result["generated_text"]
+    elif isinstance(result, list) and "generated_text" in result[0]:
+        return result[0]["generated_text"]
+    elif "text" in result:
+        return result["text"]
+    else:
+        return "⚠️ No output generated."