Hello, this is Playtoon.

We would like to address the recent issues regarding degraded output quality and response delays in some chats, and we sincerely apologize for any inconvenience caused.

Currently, the status of the model (Gemini) used in certain sections is generally unstable. Please be advised that there are aspects of this issue that are difficult to control entirely within the platform itself.

However, we take the disruption to your user experience very seriously and are responding with the following measures:

1. Prioritizing Stable Models
As of now, we are prioritizing the recommendation of models that have secured relatively stable output performance.

2. Improving Output Status Monitoring
To assist with this, you can now check the response delay status along with the average output length via the 'Notification Tab (top right of Home) > Output Status Page.'
While output results may vary depending on conversation context or model status, we will continue to improve the system so that the basic response environment can be monitored more transparently. This update will be implemented before 9:00 PM.

3. Improving Response Directionality
We are continuously working to address characteristics where certain models excessively lead the conversation or develop plots contrary to the user's flow.
As changes to response settings can affect the overall user experience, we are adjusting them carefully. We will provide a separate notice regarding the improvements and application direction as quickly as possible, within the next 3 days.

4. Testing and Integrating New Models
To reduce dependency on specific models and enhance stability, we plan to actively test and introduce new models, including the recently integrated Gelato model.
Further details regarding this will be provided in a separate announcement within next week.

We will continue to conduct rapid inspections and implement improvements to provide a more stable user experience.

Thank you.