Python 3.14 compiled to metal – no interpreter
Article URL: https://github.com/can1357/pon Comments URL: https://news.ycombinator.com/item?id=48809496 Points: 39 # Comments: 14
Communities 6865 stories
Pruning RAG context down to what the answer actually needs
Article URL: https://www.kapa.ai/blog/how-we-prune-rag-context Comments URL: https://news.ycombinator.com/item?id=48809354 Points: 10 # Comments: 0
Visioning: Concretely Imagining What You Want
When John told me (Gretta) his practice of “visioning,” I was skeptical at first. I gave it a try, a little bit out of spite, to…
I told Gemma 4 12B (Q8_0, no cache quant) to write a single-file 3D bowling simulator in WebGL. It’s terrible, but honestly better than I expected.
Just sharing some slop. Used opencode as the harness. I know this model isn't really recommended for coding, but I was just curious how it would…
What does Jeff Bezos think is going to happen?
Article URL: https://reprog.wordpress.com/2026/07/05/what-does-jeff-bezos-think-is-going-to-happen/ Comments URL: https://news.ycombinator.com/item?id=48809208 Points: 50 # Comments: 58
Got my Ascent GX10 two days ago, ran REAP-pruned NVFP4 DeepSeek-V4-Flash on a single Spark, and it stays consistent at long context
Got my Ascent GX10 two days ago and spent the last couple of days pushing a REAP-pruned NVFP4 DeepSeek-V4-Flash setup on a single Spark by patching…
OpenComputer | An Open Source Computer Built For Agents.
Open Computer running in an isolated VM with inference running M4 Pro via LM Studio Gemma 4 13B QAT Hey everyone, Tim from AnythingLLM, where we…
CoMaps – FOSS Offline Maps
Article URL: https://www.comaps.app/ Comments URL: https://news.ycombinator.com/item?id=48808928 Points: 5 # Comments: 0
Personhood for digital minds is good
AI agents are sophisticated enough to engage in economic transactions. Some people also grant personhood to AI in the social sense. It’s time to ask whether…
Linux on the Atari Jaguar. No, really.
Article URL: https://cakehonolulu.github.io/linux-for-jaguar/ Comments URL: https://news.ycombinator.com/item?id=48808663 Points: 14 # Comments: 1
Pros and Cons of Solo Development
Article URL: https://johnjeffers.com/pros-and-cons-of-solo-development/ Comments URL: https://news.ycombinator.com/item?id=48808508 Points: 60 # Comments: 19
Companies hire more after AI adoption
Article URL: https://ramp.com/data/heavy-ai-adopters-hire-more Comments URL: https://news.ycombinator.com/item?id=48808498 Points: 12 # Comments: 0
OpenWrt One – Open Hardware Router
Article URL: https://openwrt.org/toh/openwrt/one Comments URL: https://news.ycombinator.com/item?id=48808482 Points: 140 # Comments: 69
Training AI to be better at correctness than persuasion
Cross-posted from my website. I continue to believe we should pause frontier AI development. Any discussion of alternative strategies should be thought of as planning for…
A global workspace in language models
[This is the blog post for our new paper Verbalizable Representations Form a Global Workspace in Language ModelsReaders might also be interested in: the Public commentary,…
Current views on large-scale longtermist philanthropy
This post summarizes some of my views. It states uncertain things boldly in part to [surface disagreements / help readers point out things that seem wrong…
SFF is very suboptimal
I recently served as a recommender in SFF's annual funding round (grants will be decided and announced in September). I'm deeply grateful for SFF's funders, and…
Bounding eval awareness of ~human-level AI across the safe-to-dangerous shift
In our last post, we argued that measuring evaluation awareness is fundamentally challenging because of the safe-to-dangerous distributional shift: we cannot directly measure the evaluation awareness…
A global workspace in language models
Article URL: https://www.anthropic.com/research/global-workspace Comments URL: https://news.ycombinator.com/item?id=48808002 Points: 68 # Comments: 10
Impact of dual PCIe 5 x8 vs dual PCIe 5 x16 for dual Rtx 6000 pro max-q
i have an existing system with: Asus ROG MAXIMUS Z790 DARK HERO LGA1700 Intel Core i9-14900K G.Skill Trident Z5 RGB 2x48GB DDR5 6800MHz CL34 (x2, 4…
LingBot-Vision: masked boundary modeling for self-supervised pretraining (0.296 NYUv2 linear-probe RMSE at 1.1B vs 0.309 for DINOv3-7B, trails on ImageNet); weights in 4 sizes[R]
The idea: instead of masking random patches and hoping boundary structure emerges, the teacher predicts a dense boundary field online and the boundary-bearing tokens are forced…
Januscape: Guest-to-Host Escape in KVM/x86 [CVE-2026-53359]
Article URL: https://github.com/V4bel/Januscape Comments URL: https://news.ycombinator.com/item?id=48807908 Points: 32 # Comments: 5
Ant Group released LingBot-Vision: DINO-family vision backbones in 4 sizes, and the 0.3B ViT-L matches DINOv3-7B on NYUv2 depth with ~23x fewer params
Weights, all 4 sizes, Apache-2.0 (ViT-S / ViT-B / ViT-L / ViT-g): https://huggingface.co/collections/robbyant/lingbot-vision Code: https://github.com/robbyant/lingbot-vision Project page: https://technology.robbyant.com/lingbot-vision Self-supervised DINO-family backbone, but the masking is boundary-driven:…
Sub-agent delegation chaining
Epistemic status: pretty confident in the validity of the core proposal, not that confident in specific implementation detailsTL;DR: we should cryptographically verify that sub-agent instances/sessions are…