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

Gemini 3 Deep Think

blog.google

Post reply on HN
Print view View original

541–550 of 722 posts

Re: Gemini 3 Deep Think

#541

Post by dgb23 »

BeetleB wrote: post #269
dakolli wrote: post #74

Earlier quoted context omitted.

Its really weird how you all are begging to be replaced by llms, you think if agentic workflows get good enough you're going to keep your job? Or not have your salary reduced by 50%? If Agents get good enough it's not going to build some profitable startup for you (or whatever people think they're doing with the llm slot machines) because that implies that anyone else with access to that agent can just copy you, its…

> Its really weird how you all are begging to be replaced by llms, you think if agentic workflows get good enough you're going to keep your job? Or not have your salary reduced by 50%? The computer industry (including SW) has been in the business of replacing jobs for decades - since the 70's. It's only fitting that SW engineers finally become the target.

Is that really true? Software created an incredible amount of new types of jobs and markets.

Re: Gemini 3 Deep Think

#542

Post by cubefox »

janalsncm wrote: post #363
aliston wrote: post #169

Earlier quoted context omitted.

I'm having trouble just keeping track of all these different types of models. Is "Gemini 3 Deep Think" even technically a model? From what I've gathered, it is built on top of Gemini 3 Pro, and appears to be adding specific thinking capabilities, more akin to adding subagents than a truly new foundational model like Opus 4.6. Also, I don't understand the comments about Google being behind in agentic workflows. I know…

The term “model” is one of those super overloaded terms. Depending on the conversation it can mean: - a product (most accurate here imo) - a specific set of weights in a neural net - a general architecture or family of architectures (BERT models) So while you could argue this is a “model” in the broadest sense of the term, it’s probably more descriptive to call it a product. Similarly we call LLMs “language” models e…

I'm pretty sure only the second is properly called a model, and "BERT models" are simply models with the BERT architecture.

Re: Gemini 3 Deep Think

#543

Post by blinding-streak »

sinuhe69 wrote: post #38

I'm pretty certain that DeepMind (and all other labs) will try their frontier (and even private) models on First Proof [1]. And I wonder how Gemini Deep Think will fare. My guess is that it will get half the way on some problems. But we will have to take an absence as a failure, because nobody wants to publish a negative result, even though it's so important for scientific research. [1] https://1stproof.org/

As a non-mathematician, reading these problems feels like reading a completely foreign language.

https://arxiv.org/html/2602.05192v1

Re: Gemini 3 Deep Think

#545

Post by isqueiros »

jeanloolz wrote: post #526
jorl17 wrote: post #484

Earlier quoted context omitted.

Antigravity is an embarrassment. The models feel terrible, somehow, like they're being fed terrible system prompts. Plus the damn thing kept crashing and asking me to "restart it". What?! At least Kiro does what it says on the tin.

My experience with Antigravity is the opposite. It's the first time in over 10 years that an IDE has managed to take me out a bit out of the jetbrain suite. I did not think that was something possible as I am a hardcore jetbrain user/lover.

It's literally just vscode? I tried it the other day and I couldn't tell it apart from windsurf besides the icon in my dock

Re: Gemini 3 Deep Think

#546

Post by concats »

mrandish wrote: post #311
WarmWash wrote: post #234

Earlier quoted context omitted.

>because we can no longer find tasks that are feasible for normal humans but unsolved by AI. "Answer "I don't know" if you don't know an answer to one of the questions"

I've been surprised how difficult it is for LLMs to simply answer "I don't know." It also seems oddly difficult for them to 'right-size' the length and depth of their answers based on prior context. I either have to give it a fixed length limit or put up with exhaustive answers.

> I've been surprised how difficult it is for LLMs to simply answer "I don't know."

It's very difficult to train for that. Of course you can include a Question+Answer pair in your training data for which the answer is "I don't know" but in that case where you have a ready question you might as well include the real answer anyways, or else you're just training your LLM to be less knowledgeable than the alternative. But then, if you never have the pattern of "I don't know" in the training data it also won't show up in results, so what should you do?

If you could predict the blind spots ahead of time you'd plug them up, either with knowledge or with "idk". But nobody can predict the blind spots perfectly, so instead they become the main hallucinations.

Re: Gemini 3 Deep Think

#547

Post by storystarling »

jetter wrote: post #133

it is interesting that the video demo is generating .stl model. I run a lot of tests of LLMs generating OpenSCAD code (as I have recently launched https://modelrift.com text-to-CAD AI editor) and Gemini 3 family LLMs are actually giving the best price-to-performance ratio now. But they are very, VERY far from being able to spit out a complex OpenSCAD model in one shot. So, I had to implement a full fledged "screensho…

yes, i had the same experience. As good as LLMs are now at coding - it seems they are still far away from being useful in vision dominated engineering tasks like CAD/design. I guess it is a training data problem. Maybe world models / artificial data can help here?

Re: Gemini 3 Deep Think

#548

Post by netdur »

kaveh_h wrote: post #482
rob-wagner wrote: post #297

I’ve been using Gemini 3 Pro on a historical document archiving project for an old club. One of the guys had been working on scanning old handwritten minutes books written in German that were challenging to read (1885 through 1974). Anyways, I was getting decent results on a first pass with 50 page chunks but ended up doing 1 page at a time (accuracy probably 95%). For each page, I submit the page for a transcription…

Have you tried providing multiple pages at a time to the model? It might do better transcription as it have bigger context to work with.

Gemini 3 long context is not good as Gemini 2.5

Re: Gemini 3 Deep Think

#549

Post by vampiregrey »

So last week I tried Gemini pro 3, Opus 4.6, GLM 5, Kimi2.5 so far using Kimi2.5 yeilded the best results (in terms of cost/performance) for me in a mid size Go project. Curious to know what others think ?

Re: Gemini 3 Deep Think

#550

Post by optimalsolver »

gowld wrote: post #283
olalonde wrote: post #93

Earlier quoted context omitted.

Would be cool to have a benchmark with actually unsolved math and science questions, although I suspect models are still quite a long way from that level.

Does folding a protein count? How about increasing performance at Go?

It's worth noting that neither of those were accomplished by LLMs.
Post reply on HN