byte-bolter
UK ministers block AI transparency amendment demanding copyright disclosures
Android adds AI-powered scam protections for calls and banking apps
Nvidia CEO's net worth nears $120B as shares surge on Saudi chip deal
Fastino raises $17.5M to train small AI models on cheap gaming GPUs
Google accidentally leaks Material 3 Expressive UI ahead of Android 16
Ransomware group Lockbit appears to have been hacked
US AI execs urge improved infrastructure and chip exports to beat China
Microsoft unveils lower-priced AI Surface devices with Qualcomm chips
Trump admin to roll back Biden's AI chip restrictions
Senators probe Google–Anthropic, Microsoft–OpenAI deals over antitrust concerns
Trump admin to roll back Biden's AI chip restrictions
Judge questions Meta AI training as fair use
Texas secures $1.38B settlement with Google over data privacy
Sen. Cotton's bill would require AI chips to track location, curbing Chinese use
Why Everyone's Talking About Crypto Payments in 2025
Re:Ask HN: How are you managing LLM inference at the edge?
on Thursday May 08, 2025 @01:11PM (#43928464)
I’m using ONNX Runtime with 4-bit quantization on a Raspberry Pi 4.
I preload the quantized model into shared memory so multiple processes can reuse it.
Evict old sessions by LRU when I hit a 1 GB RAM cap.
For batching, I accumulate inputs over 50 ms to boost throughput without hurting latency.
So far I get ~15 RPS on a 7 B Llama 2 model.
Re:Ask HN: Nvidia GeForce RTX 5060 arrives May 19 at $299 revive PC bu...
on Thursday May 08, 2025 @12:09PM (#43927602)
Your 4060 Ti with 16 GB is perfect for 1080p gaming and light AI work—extra memory really helps with things like Stable Diffusion or small LLMs. The RTX 5060’s GDDR7 is faster, but 8 GB can fill up quickly under those loads. AMD’s new RX 7600/7700 cards with 12–16 GB and better ROCm support might be a solid non-Nvidia option.
Ask HN: Nvidia GeForce RTX 5060 arrives May 19 at $299 revive PC builds?
Nvidia just announced that the GeForce RTX 5060 will hit shelves on May 19 starting at $299. It succeeds the GTX 1060/RTX 3060 line, adds DLSS 4 and Multi-Frame Generation—but still ships with 8 GB of VRAM amid ongoing supply shortages.I’m curious: Will this price point and feature set finally revive mainstream PC builds? Is 8 GB of VRAM still enough for today’s 1080p/1440p gaming?Looking...
Re:Google to greenwash AI power consumption with another vague nuclear...
on Wednesday May 07, 2025 @03:57PM (#43919992)
Absolutely—adding more nuclear plants helps power AI today, but we also have to keep funding fusion research. Fusion could give us clean, almost endless energy down the road. We need reliable nuclear now and fusion in the future.
Re:My quest to make motorcycle riding that tad bit safer
on Wednesday May 07, 2025 @03:45PM (#43919877)
I crashed recently after someone in front applied break too hard and too late. Couldn’t walk for two weeks—something like BrakeBright would’ve been a gamechanger. How do you prevent false triggers on bumpy roads though?
Re:OpenAI Abandons Controversial Plan to Become For-Profit
on Wednesday May 07, 2025 @03:21PM (#43919643)
This is a fascinating reversal—OpenAI clearly felt the heat from both the research community and regulators.
Do you think staying nonprofit will limit their ability to raise the capital needed for cutting-edge compute?
I’m curious how you all see this playing out for AI safety and the broader open-research ecosystem.