Card #2113 · OpenAI Platform & APIs
Forced migration to expensive Codex models with degraded performance
Users are frustrated by the removal of older, more efficient Codex models, forcing them onto more expensive versions like GPT-5.4 and 5.5. These newer models are perceived as harder to control, requiring different prompting styles to achieve similar results, and consuming tokens much faster.
Pain score
0.18/ 1.00 · weighted product of four components
Reach0.32distinct authors (log-normalized)
Recency1.00freshness decay, half-life months
Engagement1.00upvotes + comments, normalized
Monetization1.00willingness-to-pay cues in evidence
§ Evidence — 3 verified quotes
They have unceremoniously just rug-pulled the only good model off the extension. Not only that, but now we are forced to use 5.4 or 5.5 models and they are VASTLY more expensive.
GPT-5.4 suddenly became much harder to control, while GPT-5.5 works properly again. This inevitably creates the impression that users are being pushed, indirectly, toward the newer and more expensive model.
having to completely change language and style to get similar results with no other option is incredibly frustrating.
Comments are a Pro feature — join the conversation, share what you've shipped, hear what others are building.
Upgrade to Pro