GPT-5.5 • Hacker News Board

Re: GPT-5.5

#241

Post by basilgohar » Thu, Apr 23, 2026, 7:47 PM UTC

endymi0n wrote: post #216

smartmic wrote: post #154

Earlier quoted context omitted.

Gone are the days of deterministic programming, when computers simply carried out the operator’s commands because there was no other option but to close or open the relays exactly as the circuitry dictated. Welcome to the future of AI; the future we’ve been longing for and that will truly propel us forward, because AI knows and can do things better than we do.

I had this funny moment when I realized we went full circle... "INTERCAL has many other features designed to make it even more aesthetically unpleasing to the programmer: it uses statements such as "READ OUT", "IGNORE", "FORGET", and modifiers such as "PLEASE". This last keyword provides two reasons for the program's rejection by the compiler: if "PLEASE" does not appear often enough, the program is considered insuff…

Thank you for this. I somehow never heard of this. I thoroughly enjoyed reading that and the loss of sanity it resulted in,

Re: GPT-5.5

#242

Post by the_sleaze_ » Thu, Apr 23, 2026, 7:47 PM UTC

nickvec wrote: post #228

I'm conflicted whether I should keep my Claude Max 5x subscription at this point and switch back to GPT/Codex... anyone else in a similar position? I'd rather not be paying for two AI providers and context switching between the two, though I'm having a hard time gauging if Claude Code is still the "cream of the crop" for SWE work. I haven't played around with Codex much.

I have experienced 0 friction swapping between the 2 models, in fact pitting them against eachother has resulted in the highest success rate for me so far.

Re: GPT-5.5

#243

Post by dannyw » Thu, Apr 23, 2026, 7:47 PM UTC

vthallam wrote: post #142

This model is great at long horizon tasks, and Codex now has heartbeats, so it can keep checking on things. Give it your hardest problem that would take hours with verifiable constraints, you will see how good this is:) *I work at OAI.

It's genuinely so great at long horizon tasks! GPT-5.5 solved many long-horizon frontier challenges, for the first time for an AI model we've tested, in our internal evals at Canva :) Congrats on the launch!

Re: GPT-5.5

#244

Post by aliljet » Thu, Apr 23, 2026, 7:47 PM UTC

6thbit wrote: post #176

Mythos 5.5 SWE-bench Pro 77.8%* 58.6% Terminal-bench-2.0 82.0% 82.7%* GPQA Diamond 94.6%* 93.6% H. Last Exam 56.8%* 41.4% H. Last Exam (tools) 64.7%* 52.2% BrowseComp 86.9% 84.4% (90.1% Pro)* OSWorld-Verified 79.6%* 78.7% Still far from Mythos on SWE-bench but quite comparable otherwise. Source for mythos values: https://www.anthropic.com/glasswing

Mythos is only real when it's actually available. If you're using Opus 4.7 right now, you know how incredibly nerfed the Opus autonomy is in service of perceived safety. I'm not so confident this will be as great as Anthropic wants us to believe..

Re: GPT-5.5

#245

Post by dns_snek » Thu, Apr 23, 2026, 7:47 PM UTC

ulimn wrote: post #219

energy123 wrote: post #186

Earlier quoted context omitted.

Look a cost per intelligence or cost per task instead of cost per token.

Isn't the outcome / solution for a given task non-deterministic? So can we reliably measure that?

Repetition and statistics, if you have $1000++ you didn't need anyway.

Re: GPT-5.5

#246

Post by egorfine » Thu, Apr 23, 2026, 7:47 PM UTC

> We are releasing GPT‑5.5 with our strongest set of safeguards to date

...

> we’re deploying stricter classifiers for potential cyber risk which some users may find annoying initially

So we should be expecting to not be able to check our own code for vulnerabilities, because inherently the model cannot know whether I'm feeding my code or someone else's.

Re: GPT-5.5

#247

Post by genericresponse » Thu, Apr 23, 2026, 7:48 PM UTC

ulimn wrote: post #219

energy123 wrote: post #186

Earlier quoted context omitted.

Look a cost per intelligence or cost per task instead of cost per token.

Isn't the outcome / solution for a given task non-deterministic? So can we reliably measure that?

Statistically. Do many trials and measure how often it succeeds/fails.

Re: GPT-5.5

#248

Post by nickvec » Thu, Apr 23, 2026, 7:48 PM UTC

the_sleaze_ wrote: post #242

nickvec wrote: post #228

I'm conflicted whether I should keep my Claude Max 5x subscription at this point and switch back to GPT/Codex... anyone else in a similar position? I'd rather not be paying for two AI providers and context switching between the two, though I'm having a hard time gauging if Claude Code is still the "cream of the crop" for SWE work. I haven't played around with Codex much.

I have experienced 0 friction swapping between the 2 models, in fact pitting them against eachother has resulted in the highest success rate for me so far.

Interesting. I may have to give that a shot, thanks.

Re: GPT-5.5

#249

Post by yokoprime » Thu, Apr 23, 2026, 7:48 PM UTC

energy123 wrote: post #186

mudkipdev wrote: post #160

This is 3x the price of GPT-5.1, released just 6 months ago. Is no one else alarmed by the trend? What happens when the cheaper models are deprecated/removed over time?

Look a cost per intelligence or cost per task instead of cost per token.

How do I reliably measure 1 unit of intelligence?

Re: GPT-5.5

#250

Post by dgunay » Thu, Apr 23, 2026, 7:48 PM UTC

vintagedave wrote: post #193

lostmsu wrote: post #164

Earlier quoted context omitted.

I never saw that happen in Codex so there's a good chance that OpenClaw does something wrong. My main suspicion would be that it does not pass back thinking traces.

Anecdata, but I see this in Codex all the time. It takes about two rounds before it realises it's supposed to continue.

I started seeing this a lot more with GPT 5.4. 5.3-codex is really good about patiently watching and waiting on external processes like CI, or managing other agents async. 5.4 keeps on yielding its turn to me for some reason even as it says stuff like "I'm continuing to watch and wait."