frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Ask HN: Gemini.Google.Com Changing Effort Level

1•yipinwong•9m ago
Does your Gemini.Google.Com (Gemini hereafter) lower your model's effort level to low from high/medium?

Context:

Last 2 weeks, Gemini has been lowering my effort level from high/medium to low as often as possible.

I've been wondering why I've been getting crappiest answers, and doesn't obey my commands, etc. - e.g. Read "audio" file. It replies almost instantly saying "I transcribed and read it". but no. it did not.

Constantly hallucianting, no obey, slow on many tasks, doesn't catch errors, results are made up.

I unsubbed pro plan today.

Microsoft, the FAA, and Our Soviet Indifference to System Collapse

https://www.thebignewsletter.com/p/monopoly-round-up-microsoft-the-faa
1•PaulHoule•1m ago•0 comments

Adding Fortran Support to LLDB (GSoC 2026)

https://blog.llvm.org/posts/2026-09-07-gsoc-add-fortran-support-to-lldb/
1•JDevlieghere•2m ago•0 comments

Show HN: Cppgm compiles C++ to WASM 4x faster than Clang

https://incpption.com/
1•vishvananda•3m ago•0 comments

Show HN: One cloud computer for all your agents

https://box.simular.ai/
1•taro666•4m ago•0 comments

Jeremy Schneider: Postgres 18.7 Debian and RedHat Packagers Protect You

https://ardentperf.com/2026/11/13/postgres-18-7-debian-and-redhat-packagers-protect-you/
1•bashhq_r•4m ago•0 comments

Chase Bank is now also having a meltdown

https://twitter.com/downdetector/status/2122345174290374924
2•samk3nny•5m ago•1 comments

Open-sourced 100 dev tools(MIT). Host locally instead of pasting secrets online

https://github.com/redbison-dev/redbison-tools
1•redbisondev•5m ago•0 comments

Tailcat

https://tailscale.com/tailcat
1•bigwheels•6m ago•0 comments

Muse Charm – Personal AI Agent

https://www.meta.com/muse-charm/
1•qainsights•7m ago•0 comments

Introducing Solaris

https://runway.com/news/research/introducing-solaris
2•gmays•8m ago•1 comments

Ask HN: Gemini.Google.Com Changing Effort Level

1•yipinwong•9m ago•0 comments

Fable 5.2 creates an underwater world in a browser in only 96 KB

https://www.tomshardware.com/tech-industry/artificial-intelligence/fable-5-2-creates-an-underwate...
1•MC995•9m ago•1 comments

Superlite-cloud: open-source software factory

https://contextfirst.blog/superlite-cloud-a-code-factory/
1•iheatuchriswogu•10m ago•1 comments

The race to build autonomous war robots faces a problem

https://www.businessinsider.com/making-autonomous-war-robots-challenges-self-driving-cars-dont-uk...
1•cjaackie•11m ago•0 comments

OpenAI

1•commoner_•12m ago•0 comments

Tailscale Bash Prompt

https://feeding.cloud.geek.nz/posts/tailscale-bash-prompt/
1•speckx•13m ago•0 comments

Oracle Critical Security Patch Update Pre-Release Announcement – November 2026

https://www.oracle.com/security-alerts/cspunov2026.html
2•choult•13m ago•0 comments

Commerce Department moves to curb remote access to AI chips

https://www.cbsnews.com/news/nvidia-jensen-huang-commerce-department-limit-remote-access-ai-chips/
1•ijidak•13m ago•0 comments

Rakazo: Open-source Grok bot alternative

https://github.com/elie222/rakazo
1•cobri•13m ago•0 comments

Farenheit 451 [pdf]

https://dn790000.ca.archive.org/0/items/fahrenheit451_202004/Fahrenheit%20451.pdf
1•srirach•14m ago•1 comments

Artificial

https://thuo.org/artificial.html
1•thuo•14m ago•0 comments

Show HN: View – Build native Mac Mini apps with your own coding agent

https://useview.app/
1•deofoo•14m ago•0 comments

Research as a Hobby

https://andrewsarafa.ghost.io/research-as-a-hobby/
1•Frankie1862•15m ago•0 comments

Show HN: Dope Wars, rebuilt as a fake 1998 desktop in the browser

https://dopefront.com/
1•kyleclouthier•15m ago•0 comments

Databricks AI Index

https://developers.databricks.com/ai-index
1•pwendell•17m ago•1 comments

The reports of Stack Overflow's death are greatly exaggerated

https://stackoverflow.blog/2026/10/29/the-reports-of-stack-overflow-s-death-are-greatly-exaggerated/
2•ktosobcy•17m ago•1 comments

New decision model from Liquid AI top of AA index

https://www.liquid.ai/blog/d1-2
2•armcat•17m ago•0 comments

Vstack – Agent skills and writing linters for engineering leaders

https://github.com/ivikash/vstack
1•ivikash•18m ago•0 comments

A New Record in Extremal Chess Composition

https://www.doubleai.com/research/maxwell-chess-record-63-checks
2•NeuralOracle•19m ago•0 comments

Nvidia board not considering reinstating Jensen Huang as chief executive

https://www.ft.com/content/e5275763-d7e2-4857-8ac6-6b888111986a
9•doener•19m ago•0 comments