Fixed a bug where the response from the LLM was being trimmed of content.

src/models.py

@@ -66,9 +66,7 @@ class HFLlamaChatModel:
         response = requests.post(api_url, headers=headers, json=query_payload)
         if response.status_code == 200:
             resp_json = json.loads(response.text)
-            llm_text = resp_json[0]['generated_text']
-            query_len = len(query_input)
-            llm_text = llm_text[query_len:].strip()
+            llm_text = resp_json[0]['generated_text'].strip()
             return llm_text
         else:
             error_detail = f"Error from hugging face code: {response.status_code}: {response.reason} ({response.content})"