GPT-4.5 What is your opinion?

A few hours ago, OpenAI released GPT-4.5, currently only for developers and users on the PRO plan.

"GPT-4.5 does not include reasoning, as it was designed to be a more general-purpose, innately smarter model."

https://help.openai.com/en/articles/10658365-gpt-4-5-in-chatgpt

Has anyone started testing it yet? What do you think?

(One of my considerations about the topic, is the cost)

MODEL | Input Cost (per 1M tokens) | Output Cost (per 1M tokens) | Context Window

GPT-4.5 = $75.00 | $150.00 | 128k tokens

GPT-4.o = $2.50 | $10.00 | 128k tokens

Claude 3.7 Sonnet = $3.00 | $15.00 | 200k tokens