switch to Gemini API key

app.py

@@ -6,6 +6,8 @@ import yaml
 from tools.final_answer import FinalAnswerTool
 from Gradio_UI import GradioUI
 from smolagents.agent_types import AgentImage
+import os
+from smolagents import LiteLLMModel
 
 # Import tools
 from tools.final_answer import FinalAnswerTool
@@ -42,13 +44,21 @@ def get_current_time_in_timezone(timezone: str) -> str:
 # If the agent does not answer, the model is overloaded, please use another model or the following Hugging Face Endpoint that also contains qwen2.5 coder:
 # model_id='https://pflgm2locj2t89co.us-east-1.aws.endpoints.huggingface.cloud' 
 
-model = HfApiModel(
-max_tokens=2096,
-temperature=0.5,
-# model_id='Qwen/Qwen2.5-Coder-32B-Instruct',# it is possible that this model may be overloaded
-model_id='https://pflgm2locj2t89co.us-east-1.aws.endpoints.huggingface.cloud',
-# model_id='deepseek-ai/deepseek-coder-33B-instruct',
-custom_role_conversions=None,
+# model = HfApiModel(
+# max_tokens=2096,
+# temperature=0.5,
+# # model_id='Qwen/Qwen2.5-Coder-32B-Instruct',# it is possible that this model may be overloaded
+# model_id='https://pflgm2locj2t89co.us-east-1.aws.endpoints.huggingface.cloud',
+# # model_id='deepseek-ai/deepseek-coder-33B-instruct',
+# custom_role_conversions=None,
+# )
+
+# Use LLM API directly
+model = LiteLLMModel(
+    model="gemini-1.5-pro",
+    api_key=os.environ.get("GEMINI_API_KEY"),
+    max_tokens=2096,
+    temperature=0.5
 )