GPT-4

OpenAI charged for PDF retrieval in the Assistant API despite stating the feature would be free until January 12, 2024. Developers incurred significantly inflated costs from repeated retrieval charges that contradicted official pricing claims.

Users report that GPT-4 performance has regressed, performing closer to GPT-3.5 than expected, and there is a widespread perception that the model was 'dumbed down' over time. Tasks that worked correctly in 2024 now produce incorrect or inconsistent results.

The Assistant API automatically appends the entire conversation thread and PDFs to each message up to the 128k token limit, with no user control over context window management. OpenAI is considering this feature but has provided no timeline.