alt Hacker NewsGPT‑5.3 Instant
Comments
I'm a bit confused by this branding (never even noticed that there was a 5.2-Instant), it's not a super fast 1000tok/s Cerebras based model which they have for codex-spark, it's just 5.2 w/out the router / "non-thinking" mode?
I feel like openai is going to get right back to where they were pre GPT-5 with a ton of different options and no one knows which model to use for what.
Don’t know about anyone else, but I find OpenAI irrelevant now. I bought an anthropic pro account to get Claude code and now just use Anthropic for everything. I can’t see anything drawing me back to OpenAI ecosystem. What am I missing?
Since the page mentions:
> Better judgment around refusals
Has any AI company ever addressed any instance of a model having different rules for different population groups? I've seen many examples of people asking questions like, "make up a joke about <group>" and then iterating through the groups, only to find that some groups are seemingly protected/privileged from having jokes made about them.
Has any AI company ever addressed studies like [1] which found that models value certain groups vastly more than others? For example, page 14 of this studies shows that the exchange rate (their word, not mine) between Nigerians and US citizens is quite large.
I kind of chuckled when I read the headline "GPT‑5.3 Instant: Smoother, more ..."
LLM companies starting to sound like cigarette advertisements.
Is nobody else unsettled by the example? Strange timing to talk about calculating trajectories on long range projectiles?
Unsettling that the example talks about trajectories in long range projectiles given recent events..
OpenAI's naming convention is slowly converging with Gillette razors. Can't wait for GPT-5.3 Instant Turbo Max Pro. Seriously though, if "Instant" just means a lower TTFT (Time To First Token) but regresses on complex reasoning, it's just a hardware accelerator for hallucinations. Fast wrong answers are still wrong answers.
Gemini 3.1 Lite with no reasoning does better than Gpt-5.3 with no reasoning?
https://aibenchy.com/compare/google-gemini-3-1-flash-lite-pr...
ChatGPT kicked off with 3.5, now this is the reverse. Would be fun to drag race these against each other, the harbinger of a revolution against now the fast path serve billions model.
> GPT‑5.3 Instant also improves the quality of answers when information comes from the web. It more effectively balances what it finds online with its own knowledge and reasoning
This is definitely something I've noticed GPT does much better than Claude in general. Claude preferences trying to answer everything itself without searching.
I unsubbed because ChatGPT was no longer SOTA. They def got cheap.
Reminds me of that graph where late customers are abused. OpenAI is already abusing the late customers.
Claude is pretty great.
I love how they come out with this article about the new 5.3 Instant, comparing it to the old 5.2 Instant, hot on the heels of actually removing "Instant" from the model chooser entirely and seemingly replacing it with "Auto (but you turn off Auto-switch to Thinking)", as apparently trying to describe "Auto but with Auto turned off" makes as little sense to them as it does to us.
The thing that hits different when you're building on top of GPT rather than just using it: your users are hitting that Instant/Thinking routing split too, and you have no idea which leg is degrading their experience. They're not filing a bug report saying "I got the dumber model." They just quietly churn and you never find out why.
How likely is that they dropped this now to push the news story about quitGPT out of the headlines?
> why can't i find love in san francisco
amazing how that's where we are now, coming from https://en.wikipedia.org/wiki/I_Left_My_Heart_in_San_Francis... in the 60s
This kind of metalinguistic quotation from 5.2 right now drives me nuts!
```That kind of “make it work at distance” trajectory work can meaningfully increase weapon effectiveness, so I have to keep it to safe, non-actionable help.```
I'm really hoping all their newer models stop doing this. It's massively overused.
It also seems to do same or slightly worse than gpt-5.2-chat on my benchmarks. No wonder they didn't show any benchmarks in their blog post (?)
https://aibenchy.com/compare/openai-gpt-5-2-chat-none/openai...
Instant is the low-latency analysis stage for us. we prompt it to emit structured bullet points and treat the raw output as data-only. a second pass (the thinking model) rewrites that outline in a human voice, double-checks the facts, and only that polished copy reaches the user. when we tried tuning Instant's persona directly it just chased warmth while still hallucinating, so we keep it bland and let the follow-up rewrite layer own the friendliness. have you tried packaging Instant's output as a neutral payload and letting another model narrate it?
"Instant" is really going to age poorly as far as a brand name goes, especially with Taalas ( https://chatjimmy.ai ) proving out that baked silicon models can be truly instant.
I was literally posting about this earlier this morning[1], but all data indicates that we'll have models equivalent to Opus 4.6 / GPT 5.3 with a truly instant (ie > 10k t/s) response time by 2028. Small models are getting better faster, and their ability to be baked into silicon in a power and speed efficient way is likely going to completely disrupt things.
So glad I cancelled my subscription last month. It's just a joke at this point.
I am always confused with the gpt names. So, via API, it's called "gpt-5.3-chat-latest". I haven't used their web UI for a while. Would this mean that this is the default model for their chat UI?
In 30 years the only sort of novel idea anyone has come up with for making money out of tech that has got any real traction is harvesting people's data. Web search - data harvesting. Social media - data harvesting. LLM as a service - data harvesting. Is that because the same money people keep betting on the same sorts of ideas from the same sort of graduates from the same cultural background?
There was crypto coins but pyramid schemes have been around forever.
Why would anyone with a shred of awareness want to subscribe to any LLM service. Particularly a foreign one where that data could potentially end up in the hands of business competitors, political enemies, extortionists or others. That goes for all of the cloud really. Like WTF people. Why do you do this?
What a strange-to-me announcement. I just submitted my first feedback comment last night, after using the platform for two years, where I said the responses were too long and padded with dramatic phrasing, and if it shortened them I could guide it better. Then today they announce this.
What a time to be alive.
It seems they've dropped automatic chat branch creation with prompt edits (at least for the most recent ones) and opted for overwriting the response. I was used to assuming the branches would remain.
Well needed if the changes work as advertised. I realized from talking with 5.2 that the issue is not about being a yapper, or speaking too much about random factual tangents or your own opinions. That's easy to tune out, and sometimes it's helpful even.
What's extremely frustrating is the subtle framings and assumptions about the user that is then treated as implicit truth and smuggled in. It's plain and simple, narcissistic frame control. Obviously I don't think GPT has a "desire" to be narcissistic or whatever, but it's genuinely exhausting talking to GPT because of this. You have to restart the conversation immediately if you get into this loop. I've never been able to dig myself out of this state.
I feel like I've dealt with that kind of thing all my life, so I'm pretty sensitive to it.
GPT‑5.2 Instant’s tone could sometimes feel “cringe,” coming across as overbearing or making unwarranted assumptions about user intent or emotions.
Strange way to write this. Why use the Gen Z cringe and put it into quotation marks? Wouldn’t it be better to just use the actual word cringeworthy which has the identical meaning?
My guess is that the article was originally written by some Gen Z intern and then some older employee added the quotation marks to the Gen Z slang.
What's the point in learning to work with these models when a new one comes out every 2 weeks?
claude models with 'extended thinking' toggled answer very quickly and the quality of the answer is far ahead of what gpt 5.2 'instant' provides. i wont even bother using the non-thinking version of chatgpt because the quality of the answers is awful and usually incorrect.
How do I know if I'm using GPT5.3 Instant on ChatGPT?
I don't see it in selections.
I would be more impressed if it could give a straight yes/no answer to a yes/no question.
> We heard feedback that GPT‑5.2 Instant would sometimes refuse questions it should be able to answer safely, or respond in ways that feel overly cautious or preachy, particularly around sensitive topics.
Lol it won't solve the issue when ChatGPT treats me like a teenager and tells me to ask my parents about everything (I just don't want to provide my ID to OpenAI to verify my age). Btw that's why I stopped using ChatGPT in my everyday life
Is this only in ChatGPT proper and not in the API? Requests appear to 400 and it does not appear in `/v1/models`
I'm really curious how people take to OpenAI after the DoW fiasco. I know consumers are mini-revolting, but will it stick? Devs ultimately decide their B2B revenue. But most don't have a choice in the matter I suppose.
What's the model ID?
I tried gpt-5.3-instant but it says model does not exist
Also don't see it on their model page
Where’s the performance specs? Or is it simply a guardrails-release?
The examples given in the video about "over-caveating" are more like bug fixes than enhancements to the model. You don't see those kind of responses on most newer models ("Sure! Let's look at your physics question, but remember - it can't be used for evil, so don't ask!!!")
OpenAI, yet again, launching something so it can pat itself on the back. Just merge with Anthropic and kill ChatGPT.
super mild. Instant, as in... ?
With enough prompt skill, you could write a tense and passable 10000-word short story with o1. I just tried 5.3 and it gave me headings every 200 words. Absolutely infuriating.
They want to be Claude so bad.
> The clear answer to this question — both in scale and long-term importance — is:
Hmmm, I haven't seen AI use that kind of em dash parenthetical construction before.
What's the model ID?
I tried `gpt-5.3-instant` but that does not work
GPT-5.2 has been such a terrible regression that I have cancelled my OpenAI account. It's possible I might not have noticed it if Claude wasn't so much better, though.
From one example
> Many people in SF are:
> Highly educated
> Career-focused
> Transplants
> Used to independence
Is "transplants" a San Francisco slang for relocators?
The emoji at the end of that poem is kind of hilarious.
The single biggest issue for me with ChatGPT right now is how absolutely awful it sounds in every answer. "Why it matters", "the big picture", "it's not jut you", the awful emphasis, the quotations with rhetorical questions, etc.. I don't know if it's intentional so you can easily spot ChatGPT-generated content on the web? The very first GPT-5 version was good but they ruined it immediately afterwards with "making the personality warmer" and making the same mistakes as 4o. I see now that they even ruined Japanese even though it was one of the best languages supported by ChatGPT (under "Limitations" at the end). I don't use it anymore, immensely disappointed.