frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Is it just me, or Opus 4.6 is sounding bit dumb lately

5•rambrrest•1h ago
Going round and round in one of the harness I use.

Comments

Imustaskforhelp•1h ago
Not using opus 4.6 but I have heard the same things.

https://old.reddit.com/r/LocalLLaMA/comments/1sgd7fp/its_ins...

There was also another post about how the perceived qualities of these models is going insanely down, something not reflected in benchmarks

I feel like it might be because the costs of GPU is reflecting back up and they might be having a more diluted model which makes it more dumb while still taking the 100$

I personally feel like this theory of these models slowing going down in intelligence until a new model which isn't bogged down intentionally might be of more interest than people think because my experience with even claude sonnet 3.7 when it had first launched was genuinely fascinating and gemini 3.1 premium and it really aligns with my personal experience tinkering with these models.

The AI industry feels quite scam-my to be honest and we would all be forced by IPO or index funds bending backwards to be left holding the bags :-/

It really feels like a great deception being played against the masses.

Areena_28•1h ago
Yeah, i noticed it too. Something feels off with the reasoning on complex multi-step tasks compared to a few months ago. hard to tell if it's actual regression or just expectations creeping up as you use it more.

Been mixing Opus with Sonnet depending on the task. Sonnet handles most things well enough and Opus for anything that genuinely needs deeper reasoning. Try it out, may be you find it useful

uberman•14m ago
Isn't it the expected thing that LLMs degrade over time?

Library Might Do Less

https://martiansoftware.com/articles/the-best-library-might-do-less
1•Tomte•1m ago•0 comments

Meta's new model is Muse Spark, and meta.ai chat has some interesting tools

https://simonwillison.net/2026/Apr/8/muse-spark/
1•gmays•1m ago•0 comments

2SLGBTQQIA+ Sub-Working Group MMIWG2SLGBTQQIA+ National Action Plan Final Report

https://scholars.wlu.ca/indg_faculty/1/
1•mhb•7m ago•0 comments

How I changed my engineering teams' workflow after a year of resisting AI tools

https://shiftmag.dev/as-an-engineering-manager-i-couldnt-ignore-ai-if-my-teams-are-to-survive-9061/
1•choochilla4•9m ago•0 comments

Atanasoff-Berry Computer

https://www.computerhistory.org/revolution/birth-of-the-computer/4/99
1•widenrun•10m ago•0 comments

ClawNetwork – A blockchain built for OpenClaw and all AI agents

https://chain.clawlabz.xyz/en
1•ludis•10m ago•0 comments

Ursa – a new Iceberg-first storage engine for Kafka

https://topicpartition.io/blog/ursa-a-new-lakehouse-first-storage-engine-for-kafka
3•enether•14m ago•0 comments

Applying "Programming Without Pointers" to an mbox indexer using Zig

https://simonhartcher.com/posts/2026-04-08-applying-programming-without-pointers-to-an-mbox-index...
1•mpweiher•14m ago•0 comments

Google: MCP Toolbox

https://github.com/googleapis/mcp-toolbox
2•danielmorozoff•14m ago•0 comments

South Africa's VALR Launches AI Service for Humans and AI Agents

https://mybroadband.co.za/news/industrynews/639519-valr-launches-ai-service-for-humans-and-ai-age...
1•DistantCl3ric•14m ago•0 comments

How Censorship Seized America

https://unherd.com/2026/04/how-censorship-seized-america/
1•barry-cotter•15m ago•0 comments

Claude Scholar

https://github.com/Galaxy-Dawn/claude-scholar
1•danielmorozoff•18m ago•0 comments

The acyclic e-graph: Cranelift's mid-end optimizer

https://cfallin.org/blog/2026/04/09/aegraph/
1•tekknolagi•18m ago•0 comments

Analysis of Virtual.1 by Bagolymadar

https://farena.in/reverse%20engineering/code%20virtualization/virtual-1-vm-solution/
1•speckx•21m ago•0 comments

Show HN: CodeJourney – Verifiable due diligence for code and IP

https://github.com/adaptive-scale/codejourney
1•debarshri•21m ago•0 comments

Scientists question CIA "Ghost Murmur" long-range heartbeat detection

https://www.scientificamerican.com/article/what-is-the-quantum-ghost-murmur-purportedly-used-in-i...
1•daniel_iversen•26m ago•0 comments

Llama LLM Network Feture

https://github.com/Jeffyjefchat/oracle-protocol
1•jeffy864567•26m ago•0 comments

Google rolls out Gmail end-to-end encryption on mobile devices

https://www.bleepingcomputer.com/news/google/google-rolls-out-gmail-end-to-end-encryption-on-mobi...
1•elsewhen•27m ago•1 comments

My Picture of the Present in AI

https://www.lesswrong.com/posts/WjaGAA4xCAXeFpyWm/my-picture-of-the-present-in-ai
1•speckx•29m ago•0 comments

Get Smart with Stupid Models

https://www.drewlyton.com/story/get-smart-with-stupid-models/
2•drew_lytle•29m ago•1 comments

Show HN: LLM-wiki LLM-compiled knowledge bases with multi-agent research v0.0.20

https://llm-wiki.net/
2•nvk•30m ago•0 comments

An open letter to 21 people I know who died

https://laylahughes.substack.com/p/an-open-letter-to-21-people-i-know
2•surprisetalk•32m ago•1 comments

"Love, Mum" – What AIs can't see about abuse

https://nataliercargill.substack.com/p/love-mum-what-ais-cant-see-about
2•surprisetalk•32m ago•0 comments

Covfefe (2017)

https://en.wikipedia.org/wiki/Covfefe
4•surprisetalk•32m ago•0 comments

The Myth of the Autism Epidemic

https://newsletter.humanprogress.org/p/the-myth-of-the-autism-epidemic
2•surprisetalk•32m ago•1 comments

AI-Powered Policing

https://www.washingtonpost.com/national-security/2026/04/10/ai-police-criminal-investigations/
1•dryadin•33m ago•0 comments

The hottest Apple product isn't what you think

https://www.cnn.com/2026/04/10/tech/apple-mac-mini-ai-openclaw
1•mooreds•33m ago•0 comments

Amazon to end support for older Kindles

https://www.bbc.com/news/articles/c98k91yy4z4o
4•akeck•33m ago•0 comments

Documentation is all you need

https://blog.unlegacy.ai/documentation-is-all-you-need/
1•Absonsonson•34m ago•1 comments

Silent YouTube Premium Price Increase?

https://old.reddit.com/r/youtube/comments/1sh8zxt/silent_youtube_premium_price_increase/
1•lordleft•34m ago•0 comments