The End of Moore's Law for AI? Gemini Flash Offers a Warning

> This is the first time a major provider has backtracked on the price of an established model

Arguably that was Haiku 3.5 in October 2024.

I think the same hypothesis could apply though, that you price your model expecting a certain average input size, and then adjust price up to accommodate the reality that people use that cheapest model when they want to throw as much as they can into the context.

Haiku 3.5 was a completely different model from Haiku 3, and part of a new model generation.

Gemini Flash 2.5 and Gemini 2.5 Flash Preview were presumably a whole lot more similar to each other.

Is there an expectation Haiku 3.5 is completely different? Even leaving semantic versioning aside, even if the .5 symbolizes a "halfway point" between major releases, it still suggests a non-major release to me.

Consumers have no idea what Haiku is.

Engineers who work with LLM APIs are hopefully paying enough attention that they understand the difference between Claude 3, Claude 3.5 and Claude 4.