Bug Report: Model Misidentification on OpenAI Platform (GPT-4.1 vs GPT-4o)
`Steps to Reproduce`
Summary: The OpenAI platform inconsistently identifies the selected GPT-4.1 model. Roughly 50% of sessions load GPT-4.1 correctly, while the rest appear to run on GPT-4o. Environment: Platform: OpenAI Developer Environment Selected Model: GPT-4.1 Occurrence: ~50% of sessions time: since three weeks Issue Description: When starting new sessions with GPT-4.1, behavior alternates between GPT-4.1 and GPT-4o. The difference is evident in: System prompt interpretation Response structure and reasoning
`Expected Result`
Request: Please review routing for GPT-4.1 and confirm when consistent model assignment is restored.
`Actual Result`
Summary: The OpenAI platform inconsistently identifies the selected GPT-4.1 model. Roughly 50% of sessions load GPT-4.1 correctly, while the rest appear to run on GPT-4o.
`Environment`
platform, assistent, Model 4.1
proud dustBOT