Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Gemini 2.5 Deep Think

blog.google

Post reply on HN
Print view View original

181–190 of 259 posts

Re: Gemini 2.5 Deep Think

#181

Post by hnav »

holowoodman wrote: post #122
asaddhamani wrote: post #19

I find it interesting, how OpenAI came out with a $200 plan, Anthropic did $100 and $200, then Gemini ups it to $250, and now Grok is at $300. OpenAI is the only one that says "practically unlimited" and I have never hit any limit on my ChatGPT Pro plan. I hit limits on Claude Max (both plans) several times. Why are these companies not upfront about what the limits are?

Because they want to have their cake and eat it too. A fair pricing model would be token-based, so that a user can see for each query how much they cost, and only pay for what they actually used. But AI companies want a steady stream of income, and they want users to pay as much as possible, while using as little as possible. Therefore they ask for a monthly or even yearly price with an unknown number of tokens inclu…

I don't think it's that, I think they just want people to onboard onto these things before understanding what the actual cost might be once they're not subsidized by megacorps anymore. Something similar to loss-leading endeavors like Uber and Lyft in the 2010s, I suspect that that showing the actual cost of inference would raise questions about the cost effectiveness of these things for a lot of applications. Internally, Google's data query surface tell you cost in terms of SWE-time (e.g. this query cost 1 SWE hour) since the incentives are different.

Re: Gemini 2.5 Deep Think

#182

Post by teitoklien »

holowoodman wrote: post #122
asaddhamani wrote: post #19

I find it interesting, how OpenAI came out with a $200 plan, Anthropic did $100 and $200, then Gemini ups it to $250, and now Grok is at $300. OpenAI is the only one that says "practically unlimited" and I have never hit any limit on my ChatGPT Pro plan. I hit limits on Claude Max (both plans) several times. Why are these companies not upfront about what the limits are?

Because they want to have their cake and eat it too. A fair pricing model would be token-based, so that a user can see for each query how much they cost, and only pay for what they actually used. But AI companies want a steady stream of income, and they want users to pay as much as possible, while using as little as possible. Therefore they ask for a monthly or even yearly price with an unknown number of tokens inclu…

you're right, about their intentions in the future. But right now, they are literally losing money every single time someone uses their product...

In most cases, atleast claude does for sure. So yea, for now, they're losing money anyways

Re: Gemini 2.5 Deep Think

#183

Post by alecco »

nusl wrote: post #57

Been using Gemini for a few months, somehow it's gotten much, much worse in that time. Hallucinations are very common, and it will argue with you when you point it out. So, don't have much confidence.

Same here. I stopped using Gemini Pro because on top of it's hard to follow verbosity it was giving contradicting answers. Things that Claude Sonnet 4 could answer.

Speaking of Sonnet, I feel like it's closing the gap to Opus. After the new quotas I started to try it before Opus and now it gets complex things right more often than not. This wasn't my experience just a couple of months ago.

Re: Gemini 2.5 Deep Think

#184

Post by Invictus0 »

teitoklien wrote: post #182
holowoodman wrote: post #122

Earlier quoted context omitted.

Because they want to have their cake and eat it too. A fair pricing model would be token-based, so that a user can see for each query how much they cost, and only pay for what they actually used. But AI companies want a steady stream of income, and they want users to pay as much as possible, while using as little as possible. Therefore they ask for a monthly or even yearly price with an unknown number of tokens inclu…

you're right, about their intentions in the future. But right now, they are literally losing money every single time someone uses their product... In most cases, atleast claude does for sure. So yea, for now, they're losing money anyways

You just completely made that up

Re: Gemini 2.5 Deep Think

#186

Post by Fade_Dance »

827a wrote: post #148
com2kid wrote: post #140

Earlier quoted context omitted.

Outside of code, the current RAG strategy is throw shit tons of unstructured text at it that has been found using vector search. Some companies are doing better, but the default rag pipelines are... kind of garbage. For example, a chat bot doing recipe work should have a RAG DB that, by default, returns entire recipes. A vector DB is actually not the solution here, any number of traditional DBs (relational or even a…

IMO: Context engineering is a fascinating topic because it starts approaching the metaphysical abstract idea of what LLMs even are. If you believe that an LLM is a digital brain, then it follows that their limitation in capabilities today are a result of their limited characteristics (namely: coherent context windows). If we increase context windows (and intelligence), we can simply pack more data into the context, a…

It's also a question of general vs specialized tools. If LLMs are being used in a limited capacity, such as retrieving recipes, then a limited environment where it only has the ability to retrieve complete recipes via RAG may be ideal in the literal sense of the word. There really is nothing better than the perfect specialized tool for a specialized job.

Re: Gemini 2.5 Deep Think

#187

Post by qingcharles »

lostmsu wrote: post #160
qingcharles wrote: post #135

Earlier quoted context omitted.

This feels like the best pelicanbike yet. The singularity might be closer than we imagine. Time for a leaderboard?

Ask and you'll receive: https://pelicans.borg.games/

Nice! Is there a way I can click on the leaderboard items so I can view them?

Re: Gemini 2.5 Deep Think

#188

Post by red75prime »

Melatonic wrote: post #177
dataviz1000 wrote: post #81

Earlier quoted context omitted.

Several years ago I thought a good litmus test for mastery of coding is not finding a solution using internet search nor getting well written questions about esoteric coding problems answered on StackOverflow. For a while, I would post a question and answer my own question after I solved the problem for posterity (or AI bots). I always loved getting the "I've been working on this for 3 days and you saved my life" com…

This has been my thought for a long time - unless there is some breakthrough in AI algo I feel like we are going to hit a "creativity wall" for coding (and some other tasks).

Any reason to think that the wall will be under the human level?

Re: Gemini 2.5 Deep Think

#189

Post by starfallg »

foundry27 wrote: post #34

I started doing some experimentation with this new Deep Think agent, and after five prompts I reached my daily usage limit. For $250 USD/mo that’s what you’ll be getting folks. It’s just bizarrely uncompetitive with o3-pro and Grok 4 Heavy. Anecdotally (from my experience) this was the one feature that enthusiasts in the AI community were interested in to justify the exorbitant price of Google’s Ultra subscription. I…

The rate limits are not because of compute performance or the lack of. It's to stop people from training their own models on the very cutting edge.

Re: Gemini 2.5 Deep Think

#190

Post by novok »

mnmatin wrote: post #59
foundry27 wrote: post #34

I started doing some experimentation with this new Deep Think agent, and after five prompts I reached my daily usage limit. For $250 USD/mo that’s what you’ll be getting folks. It’s just bizarrely uncompetitive with o3-pro and Grok 4 Heavy. Anecdotally (from my experience) this was the one feature that enthusiasts in the AI community were interested in to justify the exorbitant price of Google’s Ultra subscription. I…

They might not have been ready/optimized for production, but still wanted to release it before Aug 2 EU AI Act, this way they have 2 years for compliance. So the strategy with aggressively rate-limit for few users make sense.

wheee, great way to lock in incumbents even more or lock out the EU from startups
Post reply on HN