HNHacker News
TopNewBestAskShowJobs

Aedelon

66 karma · joined February 12, 2026

submissionscomments

CUDA Tile is the biggest GPU programming shift in 20 years

pub.towardsai.net·2 pts·Aedelon·
0

A $20/month user costs OpenAI $65 in compute. AI video is a money furnace

aedelon777.substack.com·76 pts·Aedelon·
42

Anthropic's Mythos leak: 3k files in a public CMS, and what the docs revealed

medium.com·38 pts·Aedelon·
5

Testing Nvidia's FP4: Running 70B LLMs on a Single RTX 5090 with Real Benchmarks

ai.gopubby.com·2 pts·Aedelon·
0

China's AI progress by the numbers: GLM-5 benchmarks, robotaxi, and Huawei chips

medium.com·2 pts·Aedelon·
1

AI Broke ARC-AGI-2 at 84.6% – But the Key Trick Is from 1972

ai.gopubby.com·1 pts·Aedelon·
0

Pure LLMs Score 0% on ARC-AGI-2. Why the Third Wave of AI Looks Like the First

ai.gopubby.com·1 pts·Aedelon·
0

RWKV-7 beats Llama 3.2 with 3x fewer training tokens and formally exceeds TC^0

ai.gopubby.com·2 pts·Aedelon·
3

Reasoning Models Fabricate 75% of Their Explanations (ArXiv:2505.05410)

ai.gopubby.com·4 pts·Aedelon·
0

0.1% synthetic data is enough to degrade AI models (Nature, 2024)

medium.com·2 pts·Aedelon·
3