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

551–560 of 722 posts

Re: Gemini 3 Deep Think

#551

Post by thefounder »

WiSaGaN wrote: post #489
raincole wrote: post #127

Earlier quoted context omitted.

Even before this, Gemini 3 has always felt unbelievably 'general' for me. It can beat Balatro (ante 8) with text description of the game alone[0]. Yeah, it's not an extremely difficult goal for humans, but considering: 1. It's an LLM, not something trained to play Balatro specifically 2. Most (probably >99.9%) players can't do that at the first attempt 3. I don't think there are many people who posted their Balatro p…

Yes, agentic-wise, Claude Opus is best. Complex coding is GPT-5.x. But for smartness, I always felt Gemini 3 Pro is best.

Can you give an example of smartness where Gemini is better than the other 2? I have found Gemini 3 pro the opposite of smartness on the tasks I gave him (evaluation, extraction, copy writing, judging, synthesising ) with gpt 5.2 xhigh first and opus 4.5/4.6 second. Not to mention it likes to hallucinate quite a bit .

Re: Gemini 3 Deep Think

#553

Post by e40 »

wiseowise wrote: post #158
xnx wrote: post #15

Google is absolutely running away with it. The greatest trick they ever pulled was letting people think they were behind.

Their models might be impressive, but their products absolutely suck donkey balls. I’ve given Gemini web/cli two months and ran away back to ChatGPT. Seriously, it would just COMPLETELY forget context mid dialog. When asked about improving air quality it just gave me a list of (mediocre) air purifiers without asking for any context whatsoever, and I can list thousands of conversations like that. Shopping or comparing…

And it gives incorrect answers about itself and google’s services all the time. It kept pointing me to nonexistent ui elements. At least it apologizes profusely! ffs

Re: Gemini 3 Deep Think

#554

Post by AlecSchueler »

pests wrote: post #522
olyjohn wrote: post #474

Earlier quoted context omitted.

We don't equate self awareness with consciousness. Dogs are conscious, but still bark at themselves in a mirror.

Then there is the third axis, intelligence. To continue your chain: Eurasian magpies are conscious, but also know themselves in the mirror (the "mirror self-recognition" test). But yet, something is still missing.

What's missing?

Re: Gemini 3 Deep Think

#555

Post by jonathanstrange »

wiseowise wrote: post #162
jonathanstrange wrote: post #107

Earlier quoted context omitted.

You mean AI Studio or something like that, right? Because I can't see a problem with Google's standard chat interface. All other AI offerings are confusing both regarding their intended use and their UX, though, I have to concur with that.

No projects, completely forgets context mid dialog, mediocre responses even on thinking, research got kneecapped somehow and is completely uses now, uses propaganda Russian videos as the search material (what’s wrong with you, Google?), janky on mobile, consumes GIGABYTES of RAM on web (seriously, what the fuck?). Left a couple of tabs over night, Mac is almost complete frozen because 10 tabs consumed 8 GBs of RAM do…

Fair enough. I'm always astonished how different experiences are because mine is the complete opposite. I almost solely use it for help with Go and Javascript programming and found Gemini Pro to be more useful than any other model. ChatGPT was the worst offender so far, completely useless, but Claude has also been suboptimal for my use cases.

I guess it depends a lot on what you use LLMs for and how they are prompted. For example, Gemini fails the simple "count from 1 to 200 in words" test whereas Claude does it without further questions.

Another possible explanation would be that processing time is distributed unevenly across the globe and companies stay silent about this. Maybe depending on time zones?

Re: Gemini 3 Deep Think

#556

Post by XCSme »

kenjackson wrote: post #466
XCSme wrote: post #395

Earlier quoted context omitted.

If they are knowledgeable enough and pay attention, yes. Also, if they are given enough time for the task. But the idea of automation is to make a lot fewer mistakes than a human, not just to do things faster and worse.

Actually faster and worse is a very common characterization of a LOT of automation.

That's true.

The problem is that if the automation breaks at any point, the entire system fails. And programming automations are extremely sensitive to minor errors (i.e. a missing semicolon).

AI does have an interesting feature though, it tends to self-healing in a way, when given tools access and a feedback loop. The only problem is that self-healing can incorrectly heal errors, then the final reault will be wrong in hard-to-detect ways.

So the more wuch hidden bugs there are, the nore unexpectedly the automations will perform.

I still don't trust current AI for any tasks more than data parsing/classification/translation and very strict tool usage.

I don't beleive in the full-assistant/clawdbot usage safety and reliability at this time (it might be good enough but the end of the year, but then the SWE bench should be at 100%).

Re: Gemini 3 Deep Think

#557

Post by londons_explore »

CamperBob2 wrote: post #453
londons_explore wrote: post #430

Earlier quoted context omitted.

This seems true for info not in the question - eg. "Calculate the volume of a cylinder with height 10 meters". However it is less true with info missing from the training data - ie. "I have a Diode marked UM16, what is the maximum current at 125C?"

This seems fine...? https://chatgpt.com/share/698e992b-f44c-800b-a819-f899e83da2... I don't see anything wrong with its reasoning. UM16 isn't explicitly mentioned in the data sheet, but the UM prefix is listed in the 'Device marking code' column. The model hedges its response accordingly ("If the marking is UM16 on an SMA/DO-214AC package...") and reads the graph in Fig. 1 correctly. Of course, it took 18 minutes of…

Indeed that answer is awesome. Much better than Gemini 2.5 pro which invented a 16 kilovolt diode which it just hoped would be marked "UM16".

Re: Gemini 3 Deep Think

#558

Post by zipy124 »

golem14 wrote: post #289
aeyes wrote: post #53

Earlier quoted context omitted.

https://arcprize.org/leaderboard $13.62 per task - so we need another 5-10 years for the price to run this to become reasonable? But the real question is if they just fit the model to the benchmark.

A grad student hour is probably more expensive…

Grad students are incredibly cheap? In the UK for instance their stipend is £20,780 a year...

Re: Gemini 3 Deep Think

#559

Post by lm28469 »

frde_me wrote: post #429
lm28469 wrote: post #273

Earlier quoted context omitted.

There was a test and a very useful golang error that literally explain what was wrong. The model tried implementing a solution, failed and when I pointed out the error most of them just rolled back the "solution"

What exact models were you using? And with what settings? 4.6 / 5.3 codex both with thinking / high modes?

minimax 2.5, kimi k2.5, codex 5.2, gemini 3 flash and pro, glm 4.7, devstral2 123b, etc.

Re: Gemini 3 Deep Think

#560

Post by viking123 »

thefounder wrote: post #495
lukebechtel wrote: post #4

Arc-AGI-2: 84.6% (vs 68.8% for Opus 4.6) Wow. https://blog.google/innovation-and-ai/models-and-research/ge...

Am I the only one that can’t find Gemini useful except if you want something cheap? I don’t get what was the whole code red about or all that PR. To me I see no reason to use Gemini instead of of GPT and Anthropic combo. I should add that I’ve tried it as chat bot, coding through copilot and also as part of a multi model prompt generation. Gemini was always the worst by a big margin. I see some people saying it is sm…

It's garbage really, cannot get how they get so high in benchmarks.
Post reply on HN