fix: replace deprecated gemini-2.0-flash-001 model with gemini-2.5-flash

[rogerpasky]

Problem

The model gemini-2.0-flash-001 used by the multiagent host agent has been deprecated and returns a 404 NOT_FOUND error:

This model models/gemini-2.0-flash-001 is no longer available.
Please update your code to use a newer model for the latest features and improvements.

This causes the demo UI to appear completely unresponsive — messages are sent but the agent never replies. The error is silently swallowed because process_message_threadsafe submits a coroutine via run_coroutine_threadsafe and never retrieves the future's result.

Fix

1. Updated the default model from gemini-2.0-flash-001 to gemini-2.5-flash in samples/python/hosts/multiagent/host_agent.py
2. Added error handling in demo/ui/service/server/adk_host_manager.py so that when the LLM call fails, the error is:
   • Logged to stdout
   • Returned as a visible error message in the conversation
   • The message is removed from the pending queue (preventing permanent stuck state)

Testing

Verified end-to-end: the demo UI now responds correctly to user messages after the fix.

[gemini-code-assist]

Code Review

This pull request introduces error handling when processing messages in adk_host_manager.py by wrapping the asynchronous runner loop in a try-except block and appending an error message to the conversation. It also updates the default model in host_agent.py to gemini/gemini-2.5-flash. The reviewer recommends replacing the generic print statement with proper logging using logging.exception to capture the full stack trace for better debugging.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[gemini-code-assist]

Using print for error logging is discouraged in server applications. Since this block catches any generic Exception, logging the full stack trace using logging.exception is highly recommended to aid in debugging. Importing logging inline avoids modifying the file's import section outside of the diff hunk.

        except Exception as e:
            import logging
            logging.exception('Error processing message')