#DeepSeek冲击全球算力 I have been posting for a day, and many people still have bad feelings towards deepseek.

What kind of questions are sensitive or not.

But if you really try it, deepseek is indeed ahead of other large models.

Is the best way to search for text large model AI to ask questions?

No, it is actually writing.

Traditional models, even industry-leading models such as chatgpt, have a flavor of AI when they are really written.

What is the consequence? At present, all mainstream AI models have not been trained for sentences. In fact, they are only suitable for official document writing and routine templates.

But deepseek, you can even ask any question, and it can give you a sense of reality of the text.

Compared with other models, deepseek really has a sense of intelligent thinking, while other models give me more of a summary of a large number of sentences and then select specific feedback.

So the answers given to many of the same questions will be different due to the lack of data.

Putting aside national prejudices for the time being, this time, deepseek is indeed at the forefront of the world$BTC