amrrs

7 ms·

1.
How to use open models with Claude Code(dev.nebius.com)
1 points by amrrs 2d ago|0 comments
2.
Renowned NYC Washington Square Park Chess Hustler 'Chichi' Detained by ICE(chess.com)
31 points by amrrs 14d ago|11 comments
3.
Minimax H3 Max Live – An infinite broadcast is generated by the chat(twitch.tv)
2 points by amrrs 18d ago|0 comments
4.
Who is at the frontier of terminal tasks?(fidian.ai)
3 points by amrrs 19d ago|0 comments
5.
The Hugging Face incident and the road ahead(openai.com)
335 points by amrrs 21d ago|465 comments
6.
Poolside Strikes $6B Licensing Deal with Nvidia(newcomer.co)
7 points by amrrs 27d ago|0 comments
7.
Claude Code teaching macOS to natively print to the HP Laser 1008a(cdn.kuber.studio)
130 points by amrrs 29d ago|3 comments
8.
Depth-aware light injection in TypeGPU(twitter.com)
1 points by amrrs 1mo ago|0 comments
9.
by amrrs 1mo ago
Is it kind of the stuff where the benefits outweigh the side effects? I'm just scared of the unknowns with the drug with such overwhelming positive research. Maybe, I'm a paranoid person but I'm unable to believe that such a
10.
Qwen 3.8 27B 4-bit designs an animated dolphin riding a bicycle SVG(twitter.com)
2 points by amrrs 1mo ago|0 comments
11.
Stealing Reasoning Traces from Proprietary LLM APIs [pdf](stolen-thoughts.com)
3 points by amrrs 1mo ago|1 comments
12.
Mac Mouse Fix – Make Your $10 Mouse Better Than an Apple Trackpad(github.com)
3 points by amrrs 2mo ago|0 comments
13.
Gumroad spent equal amounts on AI tokens and human payroll(twitter.com)
4 points by amrrs 2mo ago|0 comments
14.
by amrrs 2mo ago
that's a very interesting point to test new models. I speak an Indian language called "Tamil" and have always tested new models with Tamil but also have tried little bit of Arabic (quranic verses) with previous GPT image 2 an
15.
Claude Fable 5 vs. Kimi K3 vs. GPT 5.6 Sol [video](youtube.com)
1 points by amrrs 2mo ago|2 comments
16.
Linus Torvalds tells AI haters to fork off(theregister.com)
21 points by amrrs 2mo ago|3 comments
17.
Prompt Used for "A Proof of the Cycle Double Cover Conjecture" [pdf](cdn.openai.com)
2 points by amrrs 2mo ago|0 comments
18.
Christopher Nolan says younger audiences are utterly rejecting AI-generated slop(twitter.com)
9 points by amrrs 2mo ago|4 comments
19.
Ollama raises $65M, grows to nearly 9M users(techcrunch.com)
8 points by amrrs 2mo ago|7 comments
20.
by amrrs 3mo ago
As someone who's trying to get better assessments, I'm struggling to come up with objective coding tasks that evaluates all aspects of real life like planning, design choices, problem solving and context usage. From your experienc
21.
Inside Cursor's wild rise(businessinsider.com)
4 points by amrrs 3mo ago|0 comments
22.
PRC-linked influence operations are targeting AI debates in the US(openai.com)
19 points by amrrs 3mo ago|15 comments
23.
The Model Is No Longer the Bottleneck(k-dense.ai)
2 points by amrrs 3mo ago|0 comments
24.
by amrrs 3mo ago
Siri AI - looks like finally Siri is getting its due update, hopefully they ship it soon!
25.
Search every file on your Mac by meaning – all Locally(hanxiao.io)
2 points by amrrs 3mo ago|0 comments
26.
Florida sues OpenAI, Sam Altman, in lawsuit over violent incidents(techcrunch.com)
3 points by amrrs 4mo ago|1 comments
27.
Accelerating Gemma 4: faster inference with multi-token prediction drafters(blog.google)
687 points by amrrs 5mo ago|330 comments
28.
Tanstack Malicious Code was a college student with Google Antigravity(twitter.com)
6 points by amrrs 5mo ago|0 comments
29.
GPT-5.5-Pro did worse in BullshitBench(twitter.com)
4 points by amrrs 5mo ago|0 comments
30.
by amrrs 5mo ago
Honestly the problem with these is how empirical it is, how someone can reproduce this? I love when Labs go beyond traditional benchies like MMLU and friends but these kind of statements don't help much either - unless it's a prop

More ›