GPT-5.5 • Hacker News Board

Re: GPT-5.5

#41

Post by cmrdporcupine » Thu, Apr 23, 2026, 6:22 PM UTC

pixel_popping wrote: post #26

tedsanders wrote: post #19

Just as a heads up, even though GPT-5.5 is releasing today, the rollout in ChatGPT and Codex will be gradual over many hours so that we can make sure service remains stable for everyone (same as our previous launches). You may not see it right away, and if you don't, try again later in the day. We usually start with Pro/Enterprise accounts and then work our way down to Plus. We know it's slightly annoying to have to…

can't wait! Thanks guys. PS: when you drop a new model, it would be smart to reset weekly or at least session limits :)

Limits were just reset two days ago.

Re: GPT-5.5

#42

Post by 2001zhaozhao » Thu, Apr 23, 2026, 6:23 PM UTC

Pricing: $5/1M input, $30/1M output

(same input price and 20% more output price than Opus 4.7)

Re: GPT-5.5

#43

Post by ChrisGreenHeur » Thu, Apr 23, 2026, 6:23 PM UTC

karmasimida wrote: post #34

louiereederson wrote: post #27

For a 56.7 score on the Artificial Intelligence Index, GPT 5.5 used 22m output tokens. For a score of 57, Opus 4.7 used 111m output tokens. The efficiency gap is enormous. Maybe it's the difference between GB200 NVL72 and an Amazon Tranium chip?

Chips doesn’t impact output quality in this magnitude

True, but the qualifying the power played a large part. Most likely nuclear power for this high quality token efficiency.

Re: GPT-5.5

#44

Post by k2xl » Thu, Apr 23, 2026, 6:23 PM UTC

Surprised to see SWE-Bench Pro only a slight improvement (57.7% -> 58.6%) while Opus 4.7 hit 64.3%. I wonder what Anthropic is doing to achieve higher scores on this - and also what makes this test particular hard to do well in compared to Terminal Bench (which 5.5 seemed to have a big jump in)

Re: GPT-5.5

#45

Post by mh- » Thu, Apr 23, 2026, 6:23 PM UTC

stefan_ wrote: post #36

tedsanders wrote: post #19

Just as a heads up, even though GPT-5.5 is releasing today, the rollout in ChatGPT and Codex will be gradual over many hours so that we can make sure service remains stable for everyone (same as our previous launches). You may not see it right away, and if you don't, try again later in the day. We usually start with Pro/Enterprise accounts and then work our way down to Plus. We know it's slightly annoying to have to…

[flagged]

Every low-effort, thought-free comment like this further discourages people from engaging here on submissions about their employer.

Please don't.

Re: GPT-5.5

#46

Post by k2xl » Thu, Apr 23, 2026, 6:24 PM UTC

ARC-AGI 3 is missing on this list - given that the SOTA before 5.5 <1% if I recall, I wonder if this didn't make meaningful progress.

Re: GPT-5.5

#47

Post by eknkc » Thu, Apr 23, 2026, 6:24 PM UTC

ativzzz wrote: post #7

I like that they waited for opus 4.7 to come out first so they had a few days to find the benchmarks that gpt 5.5 is better at

Well anectodally, 5.4 was already better than opus 4.7 so it should not have been hard.

Re: GPT-5.5

#48

Post by ZeWaka » Thu, Apr 23, 2026, 6:24 PM UTC

astlouis44 wrote: post #9

A playable 3D dungeon arena prototype built with Codex and GPT models. Codex handled the game architecture, TypeScript/Three.js implementation, combat systems, enemy encounters, HUD feedback, and GPT‑generated environment textures. Character models, character textures, and animations were created with third-party asset-generation tools The game that this prompt generated looks pretty decent visually. A big part of th…

I personally don't think the gameplay itself is that impressive.

Re: GPT-5.5

#49

Post by vessenes » Thu, Apr 23, 2026, 6:24 PM UTC

Yay. 5.4 was a frustrating model - moments of extreme intelligence (I liked it very much for code review) - but also a sort of idiocy/literalism that made it very unsuited for prompting in a vague sense. I also found its openclaw engagement wooden and frustrating. Which didn’t matter until anthropic started charging $150 a day for opus for openclaw.

Anyway - these benchmarks look really good; I’m hopeful on the qualitative stuff.

Re: GPT-5.5

#50

Post by I_am_tiberius » Thu, Apr 23, 2026, 6:24 PM UTC

I'd really like to see improvements like these: - Some technical proof that data is never read by open ai. - Proof that no logs of my data or derived data is saved. etc...