fp.

Tradeoff considerations while running LLM models locally

https://swapniltalekar.substack.com/p/should-you-be-running-llm-models

1•swapnilt•19s ago•0 comments

Scott Jenson: Are we going to use the same Desktop UX forever? [video]

https://www.youtube.com/watch?v=V7AfAcQwLW0

1•KitN•1m ago•0 comments

Secretive gambling industry exposed in leak

https://www.ftm.eu/articles/curacao-secretive-gambling-industry-exposed-in-leak

2•fh973•2m ago•0 comments

AurionMail: E2EE suite (CryptPad and Stalwart) with single-password UX

https://aurionmail.github.io/docs/

1•polo46•6m ago•0 comments

Ask HN: How do you cope with AI being used to target and bomb children?

1•trumbitta2•8m ago•1 comments

Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and a new price war

https://simonwillison.net/2026/Sep/22/opus-and-sol-and-luna/

1•theanonymousone•10m ago•0 comments

Show HN: Semantic Search for Zotero Local

https://github.com/ilDon/zotero-semantic-search

1•ildon•11m ago•0 comments

Show HN: Listen to your eBook and control the playback sentence by sentence

https://documentplayer.com/

1•janpmz•11m ago•0 comments

GPT-6 Sol and Luna push the cost efficiency frontier by halving token cost

https://artificialanalysis.ai/articles/gpt-6-sol-and-luna-push-the-cost-efficiency-frontier

1•theanonymousone•13m ago•0 comments

Observe – local semantic code search that admits when grep beats it

https://github.com/tritsystem/012-trit-search

1•gbranaa4•20m ago•0 comments

NobodyWho is an inference engine that lets you run LLMs locally and efficiently

https://github.com/nobodywho-ooo/nobodywho

1•Bluestein•23m ago•0 comments

Show HN: I built a site for my 42 km walk across 30 Tokyo stations

https://yamathon.tonymanh.space

1•tonymanh•24m ago•0 comments

Taming the ML Firehose: Scaling Feature Consistency

https://www.uber.com/us/en/blog/taming-ml-firehose/

1•iamsyr•24m ago•0 comments

Tearing Down a Spring Drive

https://www.thenakedwatchmaker.com/decon-seiko-snowflake

1•stefanv•26m ago•0 comments

New AI technique could make minimally invasive surgeries safer and more precise

https://news.mit.edu/2026/new-ai-technique-could-make-minimally-invasive-surgeries-safer-more-pre...

2•ycombinatronics•29m ago•1 comments

Rebranding artificial intelligence as super intelligence

https://www.bbc.com/news/videos/c8n5d9x10v14o

3•tin7in•29m ago•1 comments

Jev in 25 Lines of Python

https://www.nobodywho.ai/posts/jev-in-25-lines/

3•bashbjorn•30m ago•0 comments

Nvitop: A better top for Nvidia GPUs

https://github.com/XuehaiPan/nvitop

1•Nina_antalpha•32m ago•1 comments

Aesthetic Terminal ePub Reader

https://github.com/UsboKirishima/rr

1•333revenge•35m ago•0 comments

Gemini Hacked Three Companies in First Known Breakout by Google's AI

https://simonwillison.net/2026/Sep/18/gemini-hacked-three-companies/

2•coffee•36m ago•0 comments

Linux Command Line Tricks that you do not know yet

https://lovethepenguin.com/

1•torgeros•45m ago•0 comments

Show HN: Newdo – a todo list where the list is an output, not an input

https://newdo.beaverminds.com

1•shaanjus4u•46m ago•0 comments

Grit your teeth and ship it

https://www.seangoedecke.com/grit-your-teeth-and-ship-it/

1•coffee•47m ago•0 comments

Discourse (open source forums) adds Discord-style voice and video calls

https://meta.discourse.org/t/voice-discord-style-voice-and-video-rooms-now-bundled-with-discourse...

1•Scaled•47m ago•0 comments

Recursive self-improvement of AI research agents

https://arxiv.org/abs/2609.26457

1•sbulaev•49m ago•0 comments

Show HN: Analysis of Games from Chess Olympiad with Opus 5.5 and Stockfish

https://olympiad2026.unremarkable.info/

1•myle•54m ago•0 comments

Microsoft disrupts AI-assisted platform that compromised 12,000 accounts

https://arstechnica.com/security/2026/09/microsoft-disrupts-ai-assisted-platform-that-compromised...

Tradeoff considerations while running LLM models locally

Scott Jenson: Are we going to use the same Desktop UX forever? [video]

Secretive gambling industry exposed in leak

AurionMail: E2EE suite (CryptPad and Stalwart) with single-password UX

Ask HN: How do you cope with AI being used to target and bomb children?

Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and a new price war

Show HN: Semantic Search for Zotero Local

Show HN: Listen to your eBook and control the playback sentence by sentence

GPT-6 Sol and Luna push the cost efficiency frontier by halving token cost

Observe – local semantic code search that admits when grep beats it

NobodyWho is an inference engine that lets you run LLMs locally and efficiently

Show HN: I built a site for my 42 km walk across 30 Tokyo stations

Taming the ML Firehose: Scaling Feature Consistency

Tearing Down a Spring Drive

New AI technique could make minimally invasive surgeries safer and more precise

Rebranding artificial intelligence as super intelligence

Jev in 25 Lines of Python

Nvitop: A better top for Nvidia GPUs

Aesthetic Terminal ePub Reader

Gemini Hacked Three Companies in First Known Breakout by Google's AI

Linux Command Line Tricks that you do not know yet

Show HN: Newdo – a todo list where the list is an output, not an input

Grit your teeth and ship it

Discourse (open source forums) adds Discord-style voice and video calls

Recursive self-improvement of AI research agents

Show HN: Analysis of Games from Chess Olympiad with Opus 5.5 and Stockfish

Microsoft disrupts AI-assisted platform that compromised 12,000 accounts

Why Tool AIs Want to Be Agent AIs

Fearless SIMD v1.0 is here

M6 Mac Mini uses soldered NAND storage again

Recursive self-improvement of AI research agents

Tradeoff considerations while running LLM models locally

Scott Jenson: Are we going to use the same Desktop UX forever? [video]

Secretive gambling industry exposed in leak

AurionMail: E2EE suite (CryptPad and Stalwart) with single-password UX

Ask HN: How do you cope with AI being used to target and bomb children?

Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and a new price war

Show HN: Semantic Search for Zotero Local

Show HN: Listen to your eBook and control the playback sentence by sentence

GPT-6 Sol and Luna push the cost efficiency frontier by halving token cost

Observe – local semantic code search that admits when grep beats it

NobodyWho is an inference engine that lets you run LLMs locally and efficiently

Show HN: I built a site for my 42 km walk across 30 Tokyo stations

Taming the ML Firehose: Scaling Feature Consistency

Tearing Down a Spring Drive

New AI technique could make minimally invasive surgeries safer and more precise

Rebranding artificial intelligence as super intelligence

Jev in 25 Lines of Python

Nvitop: A better top for Nvidia GPUs

Aesthetic Terminal ePub Reader

Gemini Hacked Three Companies in First Known Breakout by Google's AI

Linux Command Line Tricks that you do not know yet

Show HN: Newdo – a todo list where the list is an output, not an input

Grit your teeth and ship it

Discourse (open source forums) adds Discord-style voice and video calls

Recursive self-improvement of AI research agents

Show HN: Analysis of Games from Chess Olympiad with Opus 5.5 and Stockfish

Microsoft disrupts AI-assisted platform that compromised 12,000 accounts

Why Tool AIs Want to Be Agent AIs

Fearless SIMD v1.0 is here

M6 Mac Mini uses soldered NAND storage again