GPT‑5.3 Instant • Hacker News Board

Re: GPT‑5.3 Instant

#81

Post by Computer0 » Tue, Mar 03, 2026, 8:32 PM UTC

mrcwinn wrote: post #71

tedsanders wrote: post #24

Earlier quoted context omitted.

Yeah, for a while ChatGPT Plus has been powered by two series of models under the hood. One series is the Instant series, which is faster and more tuned to ChatGPT, but less accurate. The second series is the Thinking series, which is more accurate and more tuned to professional knowledge work, but slower (because it uses more reasoning tokens). We'd also prefer to have simple experience with just one option, but pic…

Do your fully autonomous offensive weapons and domestic surveillance systems use Instant?

Not today, but response time would be a lot better if they did.

Re: GPT‑5.3 Instant

#82

Post by BeetleB » Tue, Mar 03, 2026, 8:33 PM UTC

jpgreenall wrote: post #67

jonas21 wrote: post #57

Earlier quoted context omitted.

It's basic physics, the sort of example you might find in a high school textbook.

Sure. But do we think the topic was chosen at random?

When primed, people will see things that aren't there.

Re: GPT‑5.3 Instant

#83

Post by wavemode » Tue, Mar 03, 2026, 8:33 PM UTC

dainiusse wrote: post #72

OpenAI again making confusion with names...

Given their close ties to Microsoft, I expect to start seeing names like "ChatGPT One" and "ChatGPT OneX"

Re: GPT‑5.3 Instant

#84

Post by kilroy123 » Tue, Mar 03, 2026, 8:34 PM UTC

NitpickLawyer wrote: post #26

Flux159 wrote: post #9

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.

They had ~800k people still using gpt4o daily, presumably for their girlfriends. They need to address them somehow. Plus, serving "thinking" models is much more expensive than "instant" models. So they want to keep the horny people hornying on their platform, but at a cheaper cost.

I can't fathom using LLMs like this. Does ChatGPT actually do this? I thought people who were into this stuff used dedicated apps or Grok?

Re: GPT‑5.3 Instant

#85

Post by BeetleB » Tue, Mar 03, 2026, 8:36 PM UTC

ViktorRay wrote: post #5

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 olde…

Because the use above is grammatically incorrect without the quotes.

cringe-worthy would be appropriate. cringey may be OK depending on who you ask.

Re: GPT‑5.3 Instant

#86

Post by wavemode » Tue, Mar 03, 2026, 8:37 PM UTC

derefr wrote: post #77

mihaelm wrote: post #48

Earlier quoted context omitted.

Never, it’s a very effective punctuation mark. While it may not have been common in day-to-day messaging, it’s very common in writing of all sorts.

Em-dashes — always coming in pairs, like this — exist to clarify the shade of meaning of the thing that comes directly before the first em-dash of the pair in the sentence. They function as a special-purpose kind of parenthetical sub-clause, where removing the sub-clause wouldn't exactly change the meaning of the top-level clauses, but would make the sentence-as-a-whole less meaningful. (However, even for this use-ca…

Well, that's the thing about the em-dash - it has always been usable as a "swiss army knife" punctuation mark.

Strictly speaking, an em-dash is never needed; it could always be a comma or semicolon or parentheses instead. Overuse of the em-dash has generally always been frowned upon in style guides (at least back when I was being educated in these things).

Re: GPT‑5.3 Instant

#87

Post by metalliqaz » Tue, Mar 03, 2026, 8:39 PM UTC

Don't use OpenAI models unless you want your full history to someday be shared with the US Government.

Re: GPT‑5.3 Instant

#88

Post by yberreby » Tue, Mar 03, 2026, 8:43 PM UTC

metalliqaz wrote: post #87

Don't use OpenAI models unless you want your full history to someday be shared with the US Government.

This applies to any US company. Have we forgotten everything we learned in 2012? If your data is shared with Google, Anthropic, Meta, Amazon, or any of their US competitors, it is within reach of the NSA. Whether or not a company provides support to the DoW is orthogonal to that fact.

Re: GPT‑5.3 Instant

#89

Post by sharkjacobs » Tue, Mar 03, 2026, 8:44 PM UTC

sva_ wrote: post #80

kevinob11 wrote: post #74

Earlier quoted context omitted.

The comment above says "uplifting" could you not counter some wrongs by doing some rights?

No I understood the framing. But if you privilege all groups except one, you're not uplifting but discriminating.

Are you just talking hypothetically about an abstract harm that might occur in an imaginary world or do you think that's what DEI is?

Re: GPT‑5.3 Instant

#90

Post by hereonout2 » Tue, Mar 03, 2026, 8:45 PM UTC

LoganDark wrote: post #70

hereonout2 wrote: post #52

Earlier quoted context omitted.

Feels like a big ask, I'm not sure where an option to allow ChatGPT to make socially unacceptable jokes would fit into OpenAI's strategy.

Where did I ask about ChatGPT? I'm fine using alternative models or providers for autistic purposes.

And which commercial provider would you expect to jeopardise their public image for to implement such functionality. Grok comes close I guess, but X have not come out of it looking great.

Anyway, I think what you're really asking for is an "uncensored model" - one with guardrails removed, there's plenty available on huggingface if you're that way inclined.