Hacker News Reader

Comment by bilater | Hacker News Reader

What is this dumb article? Gemini pro 2.5 is getting talked a lot about in AI circles. AI Studio team is cracked and has been able to avoid the committee virus. Google has problems but this piece is written by someone who is not knowledgeable about this company and the space at all.

View on HN

Agree, they write it off as:

> Gemini 2.5 barely made a ripple in the broader tech conversation.

But my sense (vibes from using it) is it's the very best model right now (though I have not tried the $200 a month exclusive openai offerings)

One of the worst thing about ChatGPT / Claude is that if you're chatting for too long, they suddenly forget everything due to the limited context length.

Google seems to have understood how important a long context length is and what I absolutely love about AI Studio is that it shows how many tokens you've consumed, so that you know that you should start wrapping it up and move to a new chat.

I also haven't used the $200 exclusive model. But I think Gemini 2.5 is by far and away the best LLM I have used.

I still use Claude for code but for any general reasoning questions, or even just for Socratic dialogue purposes, Gemini has really blown me away.

It is hard to imagine any LLM being so much better that it is worth going from $0 to $200.

I love AI Studio. It's free to use and you can switch between models. Gemini 2.5 Pro's reasoning can be a little slow so if you don't need it you can switch to Gemini 2.0 Flash (which is capable of 100-200 tokens per second). You can even switch between models later. The only downside of using it over the user-facing Gemini web app is that you need to save your chats.

Reply on news.ycombinator.com