GPT-5.4 • Hacker News Board

Re: GPT-5.4

#101

Post by surgical_fire » Thu, Mar 05, 2026, 6:48 PM UTC

simianwords wrote: post #35

minimaxir wrote: post #8

The marquee feature is obviously the 1M context window, compared to the ~200k other models support with maybe an extra cost for generations beyond >200k tokens. Per the pricing page, there is no additional cost for tokens beyond 200k: https://openai.com/api/pricing/ Also per pricing, GPT-5.4 ($2.50/M input, $15/M output) is much cheaper than Opus 4.6 ($5/M input, $25/M output) and Opus has a penalty for its beta >200…

Why would some one use codex instead?

I've been using Codex for software development personally (I have a ChatGPT account), and I use Claude at work (since it is provided by my employer).

I find both Codex and Claude Opus perform at a similar level, and in some ways I actually prefer Codex (I keep hitting quota limits in Opus and have to revert back to Sonnet).

If your question is related to morality (the thing about US politics, DoD contract and so on)... I am not from the US, and I don't care about its internal politics. I also think both OpenAI and Anthropic are evil, and the world would be better if neither existed.

Re: GPT-5.4

#102

Post by glerk » Thu, Mar 05, 2026, 6:48 PM UTC

elliotbnvl wrote: post #39

denysvitali wrote: post #6

Article: https://openai.com/index/introducing-gpt-5-4/ gpt-5.4 Input: $2.50 /M tokens Cached: $0.25 /M tokens Output: $15 /M tokens --- gpt-5.4-pro Input: $30 /M tokens Output: $180 /M tokens Wtf

Looks like it's an order of magnitude off. Missprint?

Looks like fair price discovery :)

Re: GPT-5.4

#103

Post by FergusArgyll » Thu, Mar 05, 2026, 6:48 PM UTC

dpoloncsak wrote: post #53

denysvitali wrote: post #6

Article: https://openai.com/index/introducing-gpt-5-4/ gpt-5.4 Input: $2.50 /M tokens Cached: $0.25 /M tokens Output: $15 /M tokens --- gpt-5.4-pro Input: $30 /M tokens Output: $180 /M tokens Wtf

[flagged]

Maybe it's finally a bigger pretrain?

Re: GPT-5.4

#104

Post by vtail » Thu, Mar 05, 2026, 6:48 PM UTC

jcmontx wrote: post #81

5.4 vs 5.3-Codex? Which one is better for coding?

Looking at the benchmarks, 5.4 is slightly better. But it also offers "Fast" mode (at 2x usage), which - if it works and doesn't completely depletes my Pro plan - is a no brainer at the same or even slightly worse quality for more interactive development.

Re: GPT-5.4

#105

Post by cj » Thu, Mar 05, 2026, 6:49 PM UTC

paxys wrote: post #64

cj wrote: post #50

I use ChatGPT primarily for health related prompts. Looking at bloodwork, playing doctor for diagnosing minor aches/pains from weightlifting, etc. Interesting, the "Health" category seems to report worse performance compared to 5.2.

Models are being neutered for questions related to law, health etc. for liability reasons.

I'm sometimes surprised how much detail ChatGPT will go into without giving any dislaimers.

I very frequently copy/paste the same prompts into Gemini to compare, and Gemini often flat out refuses to engage while ChatGPT will happily make medical recommendations.

I also have a feeling it has to do with my account history and heavy use of project context. It feels like when ChatGPT is overloaded with too much context, it might let the guardrails sort of slide away. That's just my feeling though.

Today was particularly bad... I uploaded 2 PDFs of bloodwork and asked ChatGPT to transcribe it, and it spit out blood test results that it found in the project context from an earlier date, not the one attached to the prompt. That was weird.

Re: GPT-5.4

#106

Post by esafak » Thu, Mar 05, 2026, 6:49 PM UTC

jcmontx wrote: post #81

5.4 vs 5.3-Codex? Which one is better for coding?

For the price, it seems the latter. I'd use 5.4 to plan.

Re: GPT-5.4

#107

Post by simianwords » Thu, Mar 05, 2026, 6:49 PM UTC

surgical_fire wrote: post #101

simianwords wrote: post #35

Earlier quoted context omitted.

Why would some one use codex instead?

I've been using Codex for software development personally (I have a ChatGPT account), and I use Claude at work (since it is provided by my employer). I find both Codex and Claude Opus perform at a similar level, and in some ways I actually prefer Codex (I keep hitting quota limits in Opus and have to revert back to Sonnet). If your question is related to morality (the thing about US politics, DoD contract and so on).…

No my question was why would I use codex over gpt 5.4

Re: GPT-5.4

#108

Post by leftbehinds » Thu, Mar 05, 2026, 6:49 PM UTC

some sloppy improvements

Re: GPT-5.4

#109

Post by energy123 » Thu, Mar 05, 2026, 6:49 PM UTC

thewebguyd wrote: post #86

Aboutplants wrote: post #31

Earlier quoted context omitted.

It seems that all frontier models are basically roughly even at this point. One may be slightly better for certain things but in general I think we are approaching a real level playing field field in terms of ability.

Kind of reinforces that a model is not a moat. Products, not models, are what's going to determine who gets to stay in business or not.

Narrative violation: revenue run rates are increasing exponentially with about 50% gross margins.

Re: GPT-5.4

#110

Post by HardCodedBias » Thu, Mar 05, 2026, 6:50 PM UTC

We'll have to wait a day or two, maybe a week or two, to determine if this is more capable in coding than 5.3, which seems to be the economically valuable capability at this time.

In terms of writing and research even Gemini, with a good prompt, is close to useable. That's likely not a differentiator.