Card #1137 · OpenAI Platform & APIs
Degradation of GPT-5 model quality and instruction following
Users are experiencing a noticeable drop in model quality, specifically regarding the ability to follow complex instructions and maintain reasoning depth. The model is perceived as less consistent, more prone to ignoring prompts, and producing lower-quality code compared to its launch performance.
Pain score
0.15/ 1.00 · weighted product of four components
Reach0.32distinct authors (log-normalized)
Recency1.00freshness decay, half-life months
Engagement1.00upvotes + comments, normalized
Monetization0.50willingness-to-pay cues in evidence
§ Evidence — 4 verified quotes
The model feels less careful, less consistent, and less obedient to instructions.
the main pattern I am seeing consistently now is its clear inability to adhere or follow the instructions given
it seems that it has gotten suddenly incapable of one shotting prompts even for the simplest UI changes which is uncharacteristic of its prior performance.
the reasoning seems to last only two or three seconds before it starts answering. This behavior reminds me a lot of smaller models or models optimized to answer quickly, but with less depth.
Comments are a Pro feature — join the conversation, share what you've shipped, hear what others are building.
Upgrade to Pro