I have a custom gpt that works generally well. However, when I flip to voice mode the GPT silently reverts to a standard 4o model conversation without notice. Does anyone know why this is happening, or how to stop it? It is confusing when it happens because the bot all of a sudden seems to lose its memory, but it is because the custom gpt is gone, and there are files uploaded to the gpt definition that are needed.
#Custom GPT silently reverting to standard model GPT
woven oar
jade whaleBOT
Hi @woven oar. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
It seems like you're experiencing an issue where your custom GPT reverts to a standard GPT-4o model when switching to voice mode. Based on the information available in the knowledge sources, here are some insights that might help explain this behavior:
-
Voice Mode Limitations: The OpenAI o1 and o1-mini models, which might be part of your custom GPT setup, do not have access to certain advanced tools and features, including voice capabilities. When you switch to voice mode, the system might default to using GPT-4o, which supports voice interactions, as it is designed to handle audio input and output. This could explain why your custom GPT seems to lose its memory and reverts to a standard model. Source
-
Advanced Voice Mode: Advanced voice mode uses models like GPT-4o, which are natively multimodal and can handle real-time audio interactions. If your custom GPT is not compatible with these requirements, it might automatically switch to a model that supports voice, such as GPT-4o. Source
-
Model Switching: If you're on a paid plan, you can switch between different models, but certain features like voice might only be available with specific models like GPT-4o. This automatic switching could be a built-in feature to ensure compatibility with voice mode. Source
To address this issue, you might consider:
- Checking Model Compatibility: Ensure that your custom GPT setup is compatible with voice mode requirements. If it relies on models that do not support voice, consider adjusting your setup to use models like GPT-4o that do.
- Reviewing Model Settings: If possible, review the settings and configurations of your custom GPT to see if there are options to maintain its functionality when switching to voice mode.
If these suggestions do not resolve the issue, you might want to reach out to OpenAI support for further assistance. They could provide more specific guidance based on your custom GPT setup and the features you are using.
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
woven oar
@jade whale Are you saying you can choose the base model for a custom gpt? I do not see this.
@jade whale Are you saying you can choose the base model for a custom gpt? I do not see this.
jade whaleBOT
Let me check the knowledge base for you again. One moment! :books:
Based on the information available in the knowledge sources, when creating a custom GPT, you are indeed able to configure various settings and capabilities, but the base model for custom GPTs is currently powered by GPT-4o. This means that while you can customize the behavior, instructions, and additional capabilities of your GPT, the underlying model remains GPT-4o.
Here’s a brief overview of how you can create and configure a custom GPT:
-
Creating a GPT: You can start creating a GPT by visiting chatgpt.com/gpts/editor and selecting "Create a GPT". This process involves using the GPT Builder to help you define the tasks and capabilities of your custom GPT.
-
Configuring Your GPT: In the Configure tab, you can set detailed instructions, add images, provide additional knowledge, and enable new capabilities like web browsing or image generation. However, the base model remains GPT-4o. Source
-
Advanced Settings: You can also define custom actions and integrate third-party APIs to extend the functionality of your GPT. Source