[HN Gopher] Gemini 3.0 Pro - early tests
       ___________________________________________________________________
        
       Gemini 3.0 Pro - early tests
        
       Author : ukuina
       Score  : 148 points
       Date   : 2025-10-02 18:26 UTC (4 hours ago)
        
 (HTM) web link (twitter.com)
 (TXT) w3m dump (twitter.com)
        
       | simonw wrote:
       | I've seen a bunch of tweets like this recently, as far as I can
       | tell they're all from people using https://aistudio.google.com/
       | who got served an A/B test.
       | 
       | A few more in this genre:
       | 
       | https://x.com/cannn064/status/1973818263168852146 - "Make a SVG
       | of a PlayStation 4 controller"
       | 
       | https://x.com/cannn064/status/1973415142302830878 "Create a
       | single, self-contained HTML5 file that mimics a macOS Sonoma-
       | style desktop: translucent menu bar with live clock, magnifying
       | dock, draggable/resizable windows, and a dynamic wallpaper. No
       | external assets; use inline SVG for icons."
       | 
       | https://x.com/synthwavedd/status/1973405539708056022 "Write full
       | HTML, CSS and Javascript for a very realistic page on Apple's
       | website for the new iPhone 18"
       | 
       | I've not seen it myself so I'm not sure how confident they are
       | that it's Gemini 3.0.
        
         | ceejayoz wrote:
         | > a very realistic page on Apple's website...
         | 
         | Is this supposed to be a _good_ example?
         | 
         | It looks like something I'd put together, and you don't want me
         | doing design work.
        
         | ajcp wrote:
         | At this point until I see one run through the Pelican Benchmark
         | I can't really take a new model seriously.
        
           | diggan wrote:
           | Unfortunately, as every public benchmark, once it ends up in
           | the training sets and/or the developers aware of it, it stops
           | being effective, and I think we've started to reach that
           | point.
           | 
           | The only thing I've found to give me some sort of
           | quantitative idea of how good a new model is, is my own
           | private benchmarks. It doesn't cover everything I want to use
           | LLMs for, and only has 20-30 tests per "category", but at
           | least I'm 99% sure it isn't in the training datasets.
        
             | simonw wrote:
             | I have a few "SVG of an X riding a Y" tests that I don't
             | publish online which I run occasionally to see if a model
             | is suspiciously better at drawing a pelican riding a
             | bicycle than some other creature on some other form of
             | transport.
             | 
             | I would be _so_ entertained if I found out an AI lab had
             | wasted their time cheating on my dumb benchmark!
        
               | ajcp wrote:
               | -> I would be so entertained if I found out an AI lab had
               | wasted their time cheating on my dumb benchmark!
               | 
               | Que intro: "The gang wastes their time cheating on a dumb
               | benchmark"
        
               | mcny wrote:
               | A shower thought I just had: there must be some AI
               | training company somewhere that has injested all It is
               | always sunny in Philadelphia, not just the text but all
               | the video from all episodes somehow...
        
               | Imustaskforhelp wrote:
               | Please do let us know through your blog post if you ever
               | find AI labs to cheat on your benchmark.
               | 
               | But now I am worried that since you have shared that you
               | do SVG of an X riding a Y thing, maybe these models will
               | try to cheat on the whole SVG of X riding Y thing instead
               | of hyper focusing the pelican.
               | 
               | So now I suppose you might need to come up with an
               | entirely new thing though :)
        
               | throwup238 wrote:
               | There are so many X and Y combinations that I find it
               | hard to believe they could realistically train for a even
               | a small fraction of them. Someone has to generate the
               | graphics output for the training.
               | 
               | A duck billed platypus riding a unicycle? A man o' war
               | riding a pyrosome? A chicken riding a Quetzalcoatlus? A
               | tardigrade riding a surf board?
        
               | gnatolf wrote:
               | You're assuming that given the collection of simonw's
               | publicly available blog posts, the creativity of those
               | combinations can't be narrowed down. Simply reverse
               | engineer his brain this way and you'll get your Xs and Ys
               | ;)
        
               | throwup238 wrote:
               | I feel like that would over fit on various snakes like
               | pythons.
        
               | fragmede wrote:
               | If we accept ChatGPT telling me that there are
               | approximately 200k common nouns in English, and then we
               | square that, we get 40 billion combinations. At one
               | second per, that's ~1200 years, but then if we
               | parallelize it on a supercomputer that can do 100,000 per
               | second that would only take 3 days. Given that ChatGPT
               | was trained on _all of the Internet_ and _every book
               | written_ , I'm not sure that still seems infeasible.
        
               | throwup238 wrote:
               | It still can't satisfactorily draw a pelican on a bicycle
               | because that's either not in the training data or the
               | signal is too weak, so why would it be able to
               | satisfactorily draw every random noun-riding-noun
               | combination just because you threw a for loop at it?
               | 
               | The point is that in order to cheat on @simonw's
               | benchmark across any arbitrary combination, they'd have
               | to come up with an absurd number of human crafted input-
               | output training pairs with human produced drawings. You
               | can't just ask ChatGPT to generate every combination
               | because all it'll produce is garbage that gets a lot
               | worse the further from a pelican riding a bicycle.
               | 
               | It might work at first for the pelican and a few other
               | animals/transport combination but what does it even mean
               | for a man o' war riding a pyrosome? I asked every model I
               | have access to generate an SVG for a "man o' war riding a
               | pyrosome" and not a single one managed to draw anything
               | resembling a pyrosome. Most couldn't even produce
               | something resembling a man o' war except as a generic
               | ellipsoid-shaped jellyfish with a few tenticles.
               | 
               | Expand that to every weird noun-noun combination and it's
               | just not practical to train even a tiny fraction of them.
        
               | fragmede wrote:
               | https://chatgpt.com/share/68def5c5-8ca4-8009-bbca-
               | feabbe0651...
               | 
               | Man'o'war on a pyrosome. I don't what you expected it to
               | look like, maybe it could be more whiteish translucent
               | instead of orange, but it looks fairly reasonable to me.
               | Took a bit over a minute with the ChatGPT app.
               | 
               | Simonw's test is for the text-only output from an LLM to
               | write an SVG, not "can a multimodal AI in 2025" generate
               | a PNG. By having pictures of pelicans on bicycles in the
               | training data in PNG format, from people wanting to see
               | one, after reading his blog, there are now raster-based
               | images from an image generation model that fairly
               | convincingly look as described in the training data. Now
               | that there's PNGs of pelicans on bicycles, we would
               | expect GPT-6 to be better at generating SVGs of something
               | it's already "seen".
               | 
               | We don't know what simonw's secret combo X and Y is, nor
               | do I want to know, because that would ruin the benchmark
               | (if it isn't ruined already by virtue of him having asked
               | it). 200k nouns is definitely high though. A bit of
               | thought could cut it down to exclude concepts and lot of
               | other things. How much spare GPU capacity OpenAI has, I
               | have no idea. But if I were there, I'd want the GPUs to
               | be running as hot as the cloud provider would let me run
               | them, because they're paying per hour, not per watt, and
               | have a low-priority queue of jobs for employees to
               | generate whatever extra training data they can think of
               | on their off hours.
               | 
               | Oh and here's the pelican PNG so the other platforms can
               | crawl this comment and slurp it up.
               | 
               | https://chatgpt.com/share/68def958-3008-8009-91fa-99127fc
               | 053...
        
               | diggan wrote:
               | > I would be so entertained if I found out an AI lab had
               | wasted their time cheating on my dumb benchmark!
               | 
               | I don't think it's necessarily "cheating", it just
               | happens as they're discovering and ingesting large ranges
               | of content. A problem of public content, it's bound to be
               | included sooner or later, directly or indirectly.
               | 
               | Nice to hear you're doing some sort of contingency
               | though, and looking forward to the inevitable blog post
               | announcing the change to a different bird and vehicle :)
        
               | svachalek wrote:
               | The thing is most of the discussion about it is
               | embarrassingly bad SVGs so training on them would
               | actually hurt their performance.
        
               | reissbaker wrote:
               | I doubt they'd cheat that obviously... But "SVG of X" has
               | become common enough that I suspect most frontier labs
               | train on it, especially since the models are multimodal
               | now anyway.
               | 
               | Not that I mind; I _want_ models to be good at generating
               | SVG! Makes icons much simpler.
        
               | fragmede wrote:
               | But how would you know it's from what you would consider
               | cheating as opposed to pelicans on bicycles existing in
               | the latest training data? Obviously your blog gets fed
               | into the training set for GPT-6, as well as everyone else
               | talking about your test, so how would the comparison to a
               | secret X riding a Y tell you if an AI lab is cheating as
               | opposed to merely there being more examples in the
               | training data?
        
               | simonw wrote:
               | Mainly because if they train on the pelican on bicycle
               | SVGs from my blog they are going to get some very weird
               | looking pelicans riding some terrible looking bicycles.
        
             | ajcp wrote:
             | That's the move right there.
        
             | latemedium wrote:
             | We need to know if big AI labs are explicitly training
             | models to generate SVGs of pelicans on bicycles. I wouldn't
             | put it past them. But it would be pretty wild in they did!
        
             | londons_explore wrote:
             | As soon as you use your private tests, all the AI companies
             | vacuum up the input to use to train the next model.
             | 
             | Obviously they're only getting the question and not a
             | perfect answer, but with today's process of generating
             | hundreds of potential answers and getting another model to
             | choose the best/correct one for training, I don't think
             | that matters.
        
             | Workaccount2 wrote:
             | I honestly think people really blow out of proportion the
             | effect of "being in the training set". The internet is
             | ridden with examples of problem/solution posts that many
             | models definitely trained on, but still get wrong.
             | 
             | More important would be post training, where the labs
             | specifically train on the exact question. But it doesn't
             | seem like this is happening for most amateur benchmarks at
             | least. All the models that are good at pelican bike have
             | been good at whatever else you throw at them to SVG.
        
       | esafak wrote:
       | We can't see the code and the challenge is pedestrian. Nothing to
       | see here.
        
       | Oras wrote:
       | These tests mean nothing; I yet to see a model that is better
       | than Sonnet 4 for coding. I tried many, all of them are sub-par,
       | even with a small code base.
        
         | nnevatie wrote:
         | Well, Codex with GPT5 High wins Claude Sonnet 4.5 - this is
         | anecdotal, but I've used both extensively.
        
           | solarkraft wrote:
           | At what speed? At some point you'll have to compare to Opus.
        
         | Bolwin wrote:
         | Well yeah no surprise. You should try glm 4.6
        
       | strongpigeon wrote:
       | Google's biggest problem in my opinion (and I'm saying that as an
       | ex-googler) is that Google doesn't have a product culture. Google
       | had the tech for something like ChatGPT for a long time, but
       | couldn't come up with that product. Instead it had to rely on
       | another company showing it the way and then copy them and try to
       | out-engineer them...
       | 
       | I still think ultimately (and somewhat sadly) Google will win the
       | AI race due to its engineering talent and the sheer amount of
       | data it has (and Android integration potential).
        
         | sho_hn wrote:
         | To be fair, according to OpenAI they started ChatGPT as a
         | demo/experiment and were taken by surprise when it went viral.
         | 
         | It may well be that they also didn't have a product culture as
         | an organization, but were willing to experiment or let small
         | teams do so.
         | 
         | It's still a lesson, but maybe a different one.
         | 
         | With organizational scale it becomes harder and harder to
         | launch experiments under the brand. Red tape increases, outside
         | scrutiny increases. Retaining the ability to do that is
         | difficult.
         | 
         | Google does experiment a fair bit (including in AI, e.g.
         | NotebookLLM and its podcast feature are I think a standout
         | example of trying to see what sticks) but they also tend to try
         | to hide their experiments in developer portals nowadays, which
         | makes it difficult to get a signal from a general consumer
         | audience.
        
           | strongpigeon wrote:
           | Google is definitely good at experimenting (and yeah
           | NotebookLLM is really cool), which is a product of the
           | bottom-up culture. The lack of a consistent story with regard
           | to AI products however is a testament to the lack of product
           | vision from the top.
        
             | ajcp wrote:
             | NotebookLM came out of Google Labs though, and in
             | collaboration with outside stakeholders. I'm not sure I
             | would call it a success of "bottom-up" culture, but a well
             | realized idea from a dedicated incubator. That doesn't
             | necessarily mean the rest of the company is so empowered or
             | product oriented.
        
           | ajcp wrote:
           | -> With organizational scale it becomes harder and harder to
           | launch experiments under the brand
           | 
           | I feel like Google tried to solve for this with their
           | `withgoogle.com` domain and it just ends up being confusing
           | or worse still, frustrating when you see something awesome
           | and then nothing ever comes of it.
        
           | thereitgoes456 wrote:
           | According to Karen Hao's _Empire of AI_ , this is only half
           | accurate. And I trust what Karen Hao says a lot more.
           | 
           | OpenAI mistakenly thought Anthropic was about to launch a
           | chatbot, and ChatGPT was a scrappy, rushed-out-the-door
           | product made from an intermediate version of GPT-4, meant to
           | one-up them. Of course, they were surprised at how popular it
           | became.
        
           | dudeinhawaii wrote:
           | If I can take a slight tangent. This is what I will remember
           | OpenAI for. Not the Closed vs Open debate. They caused the
           | democratization of access to AI models. Prior to ChatGPT, I
           | would hear about these great models Deep Mind and Google were
           | developing. They'd always stay closed behind the walls of
           | Google.
           | 
           | OpenAI forced Google to release and as a result, we have all
           | of the AI tooling, integrations, and models. Meta's leaning
           | into the stolen Llama code took this further and sparked the
           | Open Source LLM revolution (in addition to the myriad
           | contributors and researchers who built on that).
           | 
           | If we had left it to Google, I suspect they'd release tooling
           | (as they did with TensorFlow) but not an LLM that might
           | compete with their core product..
        
         | byefruit wrote:
         | And even when it does copy other products, it seems to be doing
         | a terrible job of them.
         | 
         | Google's AI offering is a complete nightmare to use. Three
         | different APIs, at least two different subscriptions,
         | documentation that uses them interchangeably.
         | 
         | For Gemini's API it's often much simpler to actually pay
         | OpenRouter the 5% surchargeto BYOK than deal with it all.
         | 
         | I still can't use my Google AI Pro account with gemini-cli..
        
           | cshores wrote:
           | As of this week you can use gemini-cli with Google AI Pro
        
           | gardnr wrote:
           | Then there's the billing dashboards...
           | 
           | It's amazing how they can show useless data while completely
           | obfuscating what matters.
        
             | ur-whale wrote:
             | Yeah, the whole billing death march is what ended up making
             | me pick OpenAI as my main worhorse instead of GOOG.
             | 
             | Not enough brain cycles to figure out a way to give Google
             | money, whereas the OpenAI subscription was basically a no-
             | brainer.
        
           | specproc wrote:
           | I had great fun this week with the batch API. A good morning
           | lost trying to work out how to do a not particularly complex
           | batch request via JSONL.
           | 
           | The python library is not well documented, and has some
           | pretty basic issues that need looking at. Terrible, unhelpful
           | errors, and "oh, so this works if I put it in camel-case"
           | sort of stuff.
        
         | xnx wrote:
         | > Google doesn't have a product culture
         | 
         | Fair criticism that it took someone else to make something of
         | the tech that Google initially invented, but Google is
         | furiously experimenting with all their active products since
         | Sundar's "code red" memo.
        
         | adventured wrote:
         | Along with its engineering talent and resource scale, I think
         | their in-house chips are one of their core advantages. They can
         | scale in a way that their peers are going to struggle to match,
         | and at much lower cost. Nvidia's extreme margins are Google's
         | opportunity.
        
         | renewiltord wrote:
         | Well, they had an internal ethics team that told them that
         | their technology was garbage. That can't help. The other guys'
         | ethics teams are all like "Our stuff is too awesome for people
         | to use. No one should have this kind of unbridled power. We
         | must muzzle the beast before a tourist rides him" and Google's
         | ethics team was like "our shit sucks lol this is just a Markov
         | chain parrot doesn't do shit it's garbage".
        
           | Filligree wrote:
           | Which, to be fair--we're talking about the pre-GPT-3.5 era--
           | it kind of _was?_
        
             | renewiltord wrote:
             | The unfortunate truth when you're on the cusp of a new
             | technology: it isn't good yet. Keeping a team of guys
             | around whose sole job it is to tell you your stuff sucks is
             | probably not aligned with producing good stuff.
        
             | charcircuit wrote:
             | Don't you remember all of the scaremongering around how
             | unethical it would be to release a GPT3 model publicly.
             | 
             | Google personally reached out to someone trying to
             | reproduce GPT3 and convinced him to abandon his plan of
             | releasing it to the public.
        
               | Imustaskforhelp wrote:
               | And here we are after deepseek and the qwen models and so
               | so much more like glm 4.6 which are reaching sota of
               | sorts.
        
               | mlsu wrote:
               | There was scaremongering about releasing GPT-2.
               | 
               | GPT-2!!
        
         | thewebguyd wrote:
         | > is that Google doesn't have a product culture.
         | 
         | This is evident in Android and the pixel lineup, which could be
         | my favorite phone if not for some of the most baffling and
         | frustrating decisions that lead to a very weirdly disjointed
         | app experience (comparing to something like iOS's first party
         | tools).
         | 
         | Like removing location based reminders from google tasks, for
         | some reason? Still no apple shortcuts-like automation built-in,
         | keep can still do location based reminders but it's a notes app
         | so which am I supposed to use? Google tasks or keep? Well,
         | gemini adds reminders to google tasks and not keep if I wanted
         | to use keep primarily.
         | 
         | If they just spent some time polishing and integrating these
         | tools, and add some of their ML magic to it they'd blow Apple
         | out of the park.
         | 
         | All of Google's tech is cool and interesting, from a tech
         | standpoint but it's not well integrated for a full consumer
         | experience.
        
           | xooooogler wrote:
           | Google recently let go ALL -- EVERY SINGLE -- L3/L4/L5 UX
           | Researcher
           | 
           | https://www.thevoiceofuser.com/google-clouds-cuts-and-the-
           | bi...
           | 
           | Could it be argued that perhaps UX Research was not working
           | at all? Or that their recommendations were not being
           | incorporated? Or that things will get even worse now without
           | them?
        
             | seemaze wrote:
             | Maybe Apple should follow suit.. I jest, but I'm still
             | processing the liquid glass debacle.
        
               | thewebguyd wrote:
               | At least it's uniform. Unlike Material 3 expressive which
               | might look different depending on the app, or not be
               | implemented at all, or only half implemented in some of
               | Google's own apps even, much like with every other
               | Android redesign.
               | 
               | I get Google can't force it on all the OEMs with their
               | custom skins, but they can at least control their own
               | PixelOS and their own apps.
        
               | layer8 wrote:
               | It's not uniform at all. Some parts of the interface and
               | of their apps get it, others don't. Some parts look more
               | glassy, some more frosty. It's all over the place in
               | terms of consistency. It's also quite different between
               | Apple's various OSs, although allegedly the purpose was
               | to unify their look.
        
         | killerstorm wrote:
         | ChatGPT-3.5 was more of a novelty than a product.
         | 
         | It would be weird to release that as a serious company. They
         | tried making a deliberately-wacky chatbot but it was not fun.
         | 
         | Letting OpenAI to release it first was a right move.
        
           | Imustaskforhelp wrote:
           | To me, I want openai to release the Chatgpt 3 and chatgpt 3.5
           | as the phenomenal leap of intelligence and even I appreciated
           | the Chatgpt 3 a lot, more so than even now like It had its
           | quirks but it was such a good model man.
           | 
           | I remember forming a really simple dead simple sveltekit
           | website during Chatgpt 3. It was good, it was mind blowing
           | and I was proud of it.
           | 
           | The only interactivity was a button which would go from one
           | color to other and it would then lead to a pdf.
           | 
           | If I am going to be honest, the UI was genuinely good. It was
           | great tho and still gives me more nostalgia and good vibes
           | than current models. Em-dashes weren't that common in Chatgpt
           | 3 iirc but I have genuinely forgotten what it was like to
           | talk to it
        
         | wmf wrote:
         | Didn't Google have Bard internally around the same time as
         | ChatGPT?
        
           | eternal_braid wrote:
           | Search for Meena from Google.
        
             | gardnr wrote:
             | Most people might remember it from the headlines:
             | 
             | > In June 2022, LaMDA gained widespread attention when
             | Google engineer Blake Lemoine made claims that the chatbot
             | had become sentient. The scientific community has largely
             | rejected Lemoine's claims...
             | 
             | From https://en.wikipedia.org/wiki/LaMDA
        
               | FergusArgyll wrote:
               | Yeah, that was my introduction to LLMs!
        
             | Workaccount2 wrote:
             | https://research.google/blog/towards-a-conversational-
             | agent-...
             | 
             | Damn, that's crazy. Or at least in hindsight it is. I don't
             | remember anything big deal being made about it back then.
        
           | blueg3 wrote:
           | Bard came out shortly after ChatGPT as a prototype of what
           | would become Gemini-the-chatbot.
           | 
           | There were other, less-available prototypes prior to that.
        
         | londons_explore wrote:
         | > Android integration potential
         | 
         | Nearly all the people that matter use iPhone... Yet Apple
         | really hasn't had much success in the AI world, despite being
         | in a position to win if their product is even only vaguely
         | passable.
        
       | maerch wrote:
       | I still have a bad taste in my mouth after all those GPT-5 hype
       | articles that claimed the model was just one step away from AGI.
        
         | gardnr wrote:
         | TBF, they all believed that scaling reinforcement learning
         | would achieve the next level. They had planned to "war-dial"
         | reasoning "solutions" to generate synthetic datasets which
         | achieved "success" on complex reasoning tasks. This only really
         | produced incremental improvements at the cost of test-time
         | compute.
         | 
         | Now Grok is publicly boasting PhD level reasoning while Surge
         | AI and Scale AI are focusing on high quality datasets curated
         | by actual PhD humans.
         | 
         | Surge AI is boasting $1B in revenue, and I am wondering how
         | much of that was paid in X.ai stock:
         | https://podcasts.apple.com/us/podcast/the-startup-powering-t...
         | 
         | In my opinion the major advancements of 2025 have been more
         | efficient models. They have made smaller models much, much
         | better (including MoE models) but have failed to meaningfully
         | push the SoTA on huge models; at least when looking at the USA
         | companies.
        
           | svachalek wrote:
           | Same, qwen3 omni blows my mind for what a 30b-A3b model can
           | do. I had a video chat with it and it correctly identified
           | plant species I showed it.
        
           | ACCount37 wrote:
           | Raw model size is still pegged by the hardware.
           | 
           | You can try to build a monster the size of GPT-4.5, but even
           | if you could actually make the training stable and efficient
           | at this scale, you still would suffer trying to serve it to
           | the users.
           | 
           | Next generation of AI hardware should put them in reach, and
           | I expect that model scale would grow in lockstep with new
           | hardware becoming available.
        
       | vunderba wrote:
       | Outside of the aesthetic, the very first example on that twitter
       | post is "balls bouncing around a constrained rotating rigid
       | physics environment" which has been trivially one-shottable since
       | Claude Code was first announced.
       | 
       | It was one of the first things I tried when Claude Code went GA:
       | 
       | https://gondolaprime.pw/hex-balls
        
         | Synaesthesia wrote:
         | They have differing degrees of fidelity to the simulation, this
         | one looks pretty good and it's got parameters, but yes the
         | LLM's are really advanced now in what they can do. I was
         | actually blown away during the Gemini 2.5 announcement with
         | some of the demos people came up with.
        
       | ACCount37 wrote:
       | I hope this is the one that unfucks the multi-turn instruction
       | following.
       | 
       | One of the biggest issues holding Gemini back, IMO, compared to
       | the competitors.
       | 
       | Many LLMs are still plagued by "it's easier to reset the
       | conversation than to unfuck the conversation", but Gemini 2.5 is
       | among the worst.
        
         | solarkraft wrote:
         | Gemini's loops are a real problem. Within a few minutes of
         | using it in the CLI it happened to me me (,,I can verify that I
         | fulfilled the user's request, I can verify that I fulfilled the
         | user's request ..."). It's telling that the CLI has a detection
         | for this.
         | 
         | The other day I asked 2.5 Pro for suggestions. It would provide
         | one, which I rejected with some reasoning. It would provide
         | another, which I also rejected. Asked for more it would then
         | loop between the two, repeating the previous suggestions
         | verbatim. It went on for 3-4 times, even after being told to
         | reflect on it and it being able to recite the rejection
         | reasons.
        
       | renewiltord wrote:
       | Every three months there's some mind blowing hype around a Google
       | product, lots of people talk about it, and then when I use it
       | it's not nearly as good.
        
       | robots0only wrote:
       | In all of these posts there is someone claiming Claude is the
       | best, then somebody else claiming they have tried a bunch of
       | times and for them Gemini is the best while others find GPT-5 is
       | supreme. Obviously, all of these are subjective narrow
       | experiences. My conclusion is that all frontier models are both
       | good and bad with no clear winner and making good evals is really
       | hard.
        
         | Robdel12 wrote:
         | Yeah, my take is it's sort of up to the person using the LLM
         | and maybe how they match to that LLM. That's my hunch as to why
         | we hear wildly different takes on these LLMs working for
         | people. Gemini can be the most productive model for some while
         | others find it entirely unworkable.
        
           | jiggawatts wrote:
           | Not just personalities and preferences, but the purpose for
           | which the AI is being used also affects the results. I
           | primarily use AIs for complex troubleshooting along the lines
           | of: "Here's a megabyte of logs, an IaC template, and a
           | gibberish error code. What's the reason?" Right now, only
           | Gemini Pro 2.5 has any chance of providing a useful output
           | given those inputs, because its long-context attention is
           | better than any other model's.
        
         | binary132 wrote:
         | The fact that there is so much astroturf out there also makes
         | it difficult to evaluate these claims
        
         | SkyPuncher wrote:
         | I'll be that person:
         | 
         | * Gemini has the highest ceiling out of all of the models, but
         | has consistently struggled with token-level accuracy. In other
         | words, it's conceptual thinking it well beyond other models,
         | but it sometimes makes stupid errors when talking. This makes
         | it hard to reliably use for tool calling or structured output.
         | Gemini is also very hard to steer, so when it's wrong, it's
         | really hard to correct.
         | 
         | * Claude is extremely consistent and reliable. It's very, very
         | good at the details - but will start to forget things if things
         | get too complex. The good news is Claude is very steerable and
         | will remember those details if you remind it.
         | 
         | * GPT-5 seems to be completely random for me. It's so
         | inconsistent that it's extremely hard to use.
         | 
         | I tend to use Claude because I'm the most familiar with it and
         | I'm confident that I can get good results out of it.
        
           | Alex-Programs wrote:
           | Personally I prefer Gemini because I still use AI via chat
           | windows, and it can do a good ~90k tokens before it starts
           | getting stupid. I'm yet to find an agent that's actually
           | useful, and doesn't constantly fuck up everywhere while
           | burning money.
        
           | bcrosby95 wrote:
           | GPT-5 seems best at analyzing the codebase for me. It can
           | pick up nuances and infer strategies Claude and Gemini seem
           | to fail at.
        
           | artdigital wrote:
           | I'd say GPT-5 is the best in following and remembering
           | instructions. After an initial plan it can easily continue
           | with said plan for the next 30-60 minutes without human
           | intervention, and come back with a complete working finished
           | feature/product.
           | 
           | It's honestly crazy how good it is, coming from Claude. I
           | never thought I could already pass something a design doc and
           | have it one-shot the entire thing with such level of
           | accuracy. Even with Opus, I always need to either steer it,
           | or fix the stuff it forgot by hand / have another phase
           | afterwards to get it from 90% to 100%.
           | 
           | Yes the Codex TUI sucks but the model with high reasoning is
           | an absolute beast, and convinced me to switch from Claude Max
           | to ChatGPT Pro
        
           | Workaccount2 wrote:
           | Gemini is also the best for staying on the ball (when it
           | does) over long contexts.
           | 
           | It's really the only model that can do large(er) codebase
           | work.
        
             | brulard wrote:
             | Claude can do large code bases too, you just need to make
             | it focus on parts that matter. Most of the coding tasks
             | should not involve all parts of the code, right?
        
         | qaq wrote:
         | In my experience gemini is good at writing specs it's hit or
         | miss in reviewing code and it's not really usable for iterating
         | on code. Codex is slow but can crack issues that Claude Code
         | struggles with. So my workflow has being to use all three to
         | iterate on specs. Have claude code work on implementation and
         | have Codex review claude code's work (sometimes have gemini
         | double check it).
        
         | smoe wrote:
         | Capability wise, they seem close enough that I don't bother re-
         | evaluating them against each other all the time.
         | 
         | One advantage Gemini had (or still has, I'm not sure about the
         | other providers) was its large context window combined with the
         | ability to use PDF documents. It probably saved me weeks of
         | work on an integration with a government system uploading
         | hundreds of pages of documentation and immediately start asking
         | questions, generating rules, and troubleshooting payloads that
         | were leading to generic, computer-says-no errors.
         | 
         | No need to go trough RAG shenanigans and all of it within the
         | free token allowance.
        
         | Keyframe wrote:
         | Answer is a classic programming one - it depends? There are
         | definitely differences in strength and weaknesses among them.
         | 
         | I run claude CLI as a primary and just ask it nicely to consult
         | gemini cli (but not let it do any coding). It works
         | surprisingly well. OpenAI just fell out of my view. Even
         | cancelled ChatGPT subscription. Gemini is leaping forward and
         | _feels like_ ChatGPT-5 is a regression.. I can't put my finger
         | on it tbh.
        
         | mlsu wrote:
         | Because how good a model is is mostly just what the training
         | data is at this point.
         | 
         | It's like the personality of a person. Employee A is better at
         | talking to customers than Employee B, but Employee B is better
         | at writing code than Employee A. Is one better than the other?
         | Is one smarter than the other? Nope. Different training data.
        
       | nharada wrote:
       | Gemini has always been the leader in multimodal work like images
       | and video, I expect this won't be any different but am interested
       | to see how it is
        
       | whywhywhywhy wrote:
       | These influencer tests are so pointless and don't represent the
       | reality of model use at all when things are constantly being
       | downgraded when people actually use the thing.
       | 
       | Not to mention every team will have the bouncing balls in the
       | polygon in their dataset now.
        
       ___________________________________________________________________
       (page generated 2025-10-02 23:00 UTC)