OpenAI's new GPT-4.5 model has garnered criticism for being overpriced and underperforming, with claims that it lacks the compelling features of its predecessors. Despite being marketed as the "largest and most knowledgeable model yet", users have noted its high costs, with input and output tokens priced significantly higher than GPT-4. Critics within the AI community have labeled it as a "lemon" and a "nothingburger", indicating widespread disappointment. OpenAI emphasizes the model's emotional intelligence and conversation capabilities, but challenges such as high error rates remain prevalent.
OpenAI research scientist Mia Glease told the MIT Technology Review, "It has the ability to engage in warm, intuitive, natural, flowing conversations, and we think that it has a stronger understanding of what users mean, especially when their expectations are more implicit, leading to nuanced and thoughtful responses."
AI critic Gary Marcus characterized the LLM as a "nothingburger," reflecting the overall disappointment surrounding the GPT-4.5 model's perceived shortcomings compared to its predecessors.
The unnamed expert referred to GPT-4.5 as a "lemon," criticizing its performance relative to the significantly higher cost of $75 per million input tokens.
Despite OpenAI's claim of GPT-4.5 being their "largest and most knowledgeable model yet," expectations are tempered by concerns about its high cost, slow speed, and error rates.
Collection
[
|
...
]