Chat is showing I used 5.3 when I am using 5.4 thinking mode & skipping thinking

Understanding Model Version Discrepancies in AI Chat Platforms: A Closer Look at ChatGPT Versioning

In the realm of AI conversational tools, users often rely on precise model versions to ensure expected performance and capabilities. Recently, some users have observed confusing discrepancies in version labeling within ChatGPT, particularly concerning the transition from GPT-3.5 to GPT-4, and the associated “thinking mode.”

The Issue at Hand

A user reported that their ChatGPT interface consistently displays the model as “GPT-3.5” (or a similar earlier version), even when they are actively engaging in what should be GPT-4 “thinking mode” and utilizing skipping features designed to optimize responses. Notably, the user observed that despite claims from OpenAI representatives indicating that skipping thinking on GPT-4 still results in the GPT-4 model being used, their chats remain labeled as GPT-3.5.

Investigation and Testing

To verify these observations, the user initiated multiple new chat sessions, testing different prompts and toggles to force GPT-4 engagement. Despite these efforts, the system continued to identify the model as GPT-3.5, leading to confusion and frustration.

Implications for Users

This discrepancy highlights important considerations:

• Model Label Accuracy: The labeling of the chat model can sometimes lag behind or misrepresent the actual underlying model being used, especially when certain features such as “skipping thinking” are employed.
• Feature Functionality vs. Labeling: Features intended to optimize responses do not necessarily change the model label, but users might assume they do, which can impact expectations and trust.
• Need for Transparency: Clear communication from platform providers about what model version is actively engaged and how features impact that are essential for user confidence.

Moving Forward

For users engaging with AI chat models, it is advisable to:

• Stay updated with official announcements regarding model updates and features.
• Report inconsistencies or unexpected behaviors to platform support teams.
• Be cautious when interpreting model labels, especially if performance characteristics suggest otherwise.

Conclusion

Discrepancies between perceived and actual model versions can cause confusion, but understanding how versioning and feature toggles interact is key. Clarifying these aspects will help users better harness AI capabilities while maintaining confidence in the tools they rely on.

---

Note: If you encounter similar issues, consider reaching out to the platform’s support or checking official documentation for the latest updates on model deployment and feature functionality.