frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Ask HN: Agents get dumber before release of new model version?

6•sporkland•4h ago
I've noticed an effect with openai where my codex agents seem to perform worse in the week(s) leading up to a new release. I'm wondering if the vendors tweak effort params at all to free up hardware to host the new version. It's a double win as the new model will look night and day better to their regular users when the new model is released presumably with effort back at normal levels.

Is this a known phenomenon? Are any folks trying to measure any of this objectively?

Comments

Fizzadar•3h ago
Reminds me of a similar effect around iPhone releases (https://www.statista.com/chart/2514/iphone-releases/, many others). Can’t say I’ve noticed any change.
suprjami•33m ago
Yes, this is already a widely circulated unproven LLM conspiracy theory.

Ask HN: Favorite text heavy blogs that are a joy to read?

70•joshmarinacci•2d ago•20 comments

Ask HN: Want to build something open source on nights and weekends together?

19•vira28•14h ago•4 comments

Ask HN: How do you get into a flow state when using AI to code?

66•kilroy123•4h ago•85 comments

Ask HN: How are thinking efforts implemented?

59•simianwords•4d ago•19 comments

Ask HN: Would it be useful to have a slop button in addition to flag?

20•BugsJustFindMe•1d ago•11 comments

I procrastinate by building tools to stop me from procrastinating: A sad story

10•thisislorenzov•8h ago•5 comments

I'm Eric Ries, author of "The Lean Startup" and new book "Incorruptible" – AMA

766•eries•1d ago•543 comments

Ask HN: Agents get dumber before release of new model version?

6•sporkland•4h ago•2 comments

Notes on DeepSeek

197•vinhnx•1d ago•137 comments

AWS Bedrock to require sharing data with Anthropic for Mythos and future models

413•TomAnthony•1d ago•250 comments

Ask HN: Is there a metric for AI code quality?

3•fractalf•8h ago•2 comments

Ask HN: Are most corporate SWE jobs performative?

235•hnthrow10282910•1d ago•275 comments

Ask HN: Is anyone shorting the overspend in AI yet?

13•ggm•14h ago•12 comments

Ask HN: Why hasn't there been a real competitor to Ticketmaster yet?

259•mdni007•3d ago•232 comments

Ask HN: What internal tool did you build that became a product?

4•nehpets•15h ago•4 comments

Ask HN: Is anyone else seeing a Slack auth bug?

2•HoyaSaxa•6h ago•0 comments

Tell HN: Claude Code keeps getting worse

4•prmph•9h ago•3 comments

Ask HN: Are you still using a Vision Pro?

166•y1n0•2d ago•211 comments

Ask HN: What are tools you have made for yourself since the advent of AI?

433•aryamaan•3d ago•756 comments

Ask HN: Just me feeling that Mythos/Fabel just 1% there?

4•punnerud•14h ago•4 comments

Tell HN: np.reddit.com now redirects to www.reddit.com

4•kevinwang•3h ago•1 comments

Ask HN: Degree apprenticeships in engineering, do they exist?

2•adamofeden•3h ago•3 comments

I added a prompt to future ASI – TLBIC Policy Proposal v5 now available

2•michikawa59•11h ago•0 comments

Ask HN: What coding agents are you using?

8•linzhangrun•17h ago•13 comments

Ask HN: Temporal Awareness in LLM?

2•Pamar•12h ago•0 comments

Discussion: Fable 5 is weak at flagging prompts correctly

2•eckelhesten•12h ago•0 comments

Ask HN: Any Local LLM can I run without GPU for Local Agentic workflow AI?

6•limondas•13h ago•3 comments

Ask HN: Did Anthropic Just Win?

4•lnenad•1d ago•10 comments

Ask HN: Releasing code under AGPLv3, but want to block LLM reconstruction?

5•zionsati•19h ago•3 comments

Ask HN: What software feels exceptionally polished?

7•Adam-Hincu•16h ago•18 comments