Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance

github.com

Post reply on HN
Print view View original

31–40 of 165 posts

Re: GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance

#31

Post by sigbottle »

HumanOstrich wrote: post #26
jatora wrote: post #22

Earlier quoted context omitted.

The vibe-assumed claude code performance regression, yep. People should stop expecting consistent performance from non-deterministic systems. There is zero empirical corroboration of performance degredation. There has been a step change... in the amount of whining and complaining coders exhibit lately.

If you bother to look at the issue instead of whining and complaining, you will see the evidence.

When I disagree with the data: I will nitpick every last detail of methodology, any cross-corroboration is an anecdote, suddenly I demand a-priori levels of justification. All science is flawed anyways, it's not like mathematics, you can't get absolute certainty, so why bother? You're always going to be making base assumptions that can be challenged, you're necessarily going to abstract out the territory, the map is flawed.

When I agree with the data: I will boast about the victories of science and empiricism, we found the perfect set of natural abstractions that are necessary and sufficient to map out the territory that carve at the joints of the problem, any concern about assumptions is rebutted with generic "Well, we're just pragmatists; we're not perfect, but clearly we're converging on the right direction! You're clearly someone who just wants to nitpick and not get any work done."

My experience with certain hackernews commenters in a nutshell.

Re: GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance

#32

Post by keyle »

ghosty141 wrote: post #24

Maybe its just bad memory but I feel like 5.3 was the best version in terms of token usage and code quality. 5.5 works better but it just eviscerates tokens.

They rendered 5.3 unusable for me a few weeks back. It simply was locking up or answering poorly.

Re: GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance

#33

Post by ACCount37 »

kbdiaz wrote: post #13
kleton wrote: post #7

Clearly they are batching reasoning inference in a few multiples of 512 tokens as a throughput optimization

Isn't the standard to use continuous batching? If they are using continuous batching -- I'm curious why generated token length matters, and why they might be clustering them. If not -- I'm curious why they aren't and what is the tradeoff here.

This "~512 batching" makes me think of things like diffusion or prefill.

If they managed to put together some dirty hack that lets them generate about 512 tokens worth of reasoning in parallel instead of in sequence? That would explain it.

Re: GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance

#34

Post by tunesmith »

selectodude wrote: post #25
siva7 wrote: post #10

Earlier quoted context omitted.

I've switched 3 months ago to Codex because Claude got incredibly stupid. 6 months ago vice versa. It doesn't matter if you use Codex or Claude. Both will fuck with you at some point. Though Codex probably less.

At least OpenAI lets me use my own harness. Having to rely on insane PMs letting Claude Mythos go wild on the codebase has not been going well lately.

How do you mean OpenAI lets you use your own harness? I'm under the impression that a custom harness requires the OpenAI SDK, which requires api tokens rather than plus/pro accounts.

Re: GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance

#35

Post by thatxliner »

resonious wrote: post #18

Deja Vu... This looks just like the Claude Code performance regression back in April. I just quit my Claude subscription when that happened and went to Codex. Now I'm kinda thinking of trying per token for both, using GLM 5.2 on Fireworks for most tasks, shelling out to the big boys only when needed. Not totally confident I'll break even though.

Fireworks?

Re: GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance

#36

Post by jerezzprime »

siva7 wrote: post #10
zenapollo wrote: post #9

I’ve definitely experienced step jumps down in quality on an almost daily basis. I usually used xhigh. The experience of relying on codex’s outstandingly thorough coding earlier in the year has evaporated for me. I’m seeing incredibly stupid implementations intermittently, and have simply switched to Claude until openai takes the issue seriously. As far as i could tell they haven’t taken it seriously for the several…

I've switched 3 months ago to Codex because Claude got incredibly stupid. 6 months ago vice versa. It doesn't matter if you use Codex or Claude. Both will fuck with you at some point. Though Codex probably less.

If you use GitHub Copilot you can switch between them in the same session if you want.

Re: GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance

#37

Post by criley2 »

jatora wrote: post #23
criley2 wrote: post #15

Earlier quoted context omitted.

I'm struggling as well to understand, and I think perhaps they mean they use ChatGPT website with GPT-5.5+reasoning for problem solving, and paste the output into Codex CLI/App. I think they're saying that letting Codex CLI/App problem solve with GPT-5.5 isn't as effective. Essentially that the web harness is superior to the agentic engineering harness for problem solving? Not sure if I agree, but I do happen to use…

I assume they are lying and still think you can use gpt 5.5 non-codex within codex cli. And they outed themselves. A lot of nonsense. And the very poor communication skills just seem like the typical chinese astroturfing you see pretty often now when discussing OAI/Claude.

See, this is part of the confusion. There is no such thing as "GPT-5.5-codex". The last codex-branded model was "GPT-5.3-codex". Starting with "GPT-5.4" the main model handles agentic engineering and they did not release a coding model.

Both the web harness and codex app/cli use "GPT-5.5".

Re: GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance

#38

Post by HumanOstrich »

tunesmith wrote: post #34
selectodude wrote: post #25

Earlier quoted context omitted.

At least OpenAI lets me use my own harness. Having to rely on insane PMs letting Claude Mythos go wild on the codebase has not been going well lately.

How do you mean OpenAI lets you use your own harness? I'm under the impression that a custom harness requires the OpenAI SDK, which requires api tokens rather than plus/pro accounts.

Anthropic is the one that prohibits harnesses other than Claude Code on subscription plans and bans users for disobeying.

OpenAI officially allows that with subscriptions.

Re: GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance

#40

Post by hakunin »

tunesmith wrote: post #34
selectodude wrote: post #25

Earlier quoted context omitted.

At least OpenAI lets me use my own harness. Having to rely on insane PMs letting Claude Mythos go wild on the codebase has not been going well lately.

How do you mean OpenAI lets you use your own harness? I'm under the impression that a custom harness requires the OpenAI SDK, which requires api tokens rather than plus/pro accounts.

You must've missed the OpenAI's response to Anthropic forcing everyone to their own harness if they want subscription pricing: official endorsement of custom harnesses like opencode and pi even when used with Codex subscription.

I think they even partnered with opencode or something like that (don't remember).

Post reply on HN