Every story across every category, newest first. Each card links to the original publisher; daily-brief posts open as editorial pages.
RT DogeDesignerBREAKING: Kazakhstan Railways is bringing Starlink internet to 900 train carriages this summer.• By July, around 900 train carriages will be equipped with Starlink internet.•…
RT Dmytro Dzhulgakovyou may have heard that glm-5.2 at 280 token/s is cool, how about 318and we still have room to go
sycl : fix the failed UT cases of conv_3d (#24900) macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS Intel (x64) iOS…
Retrieving external knowledge is essential for solving real-world tasks, yet it remains challenging when the relationship between a query and its relevant knowledge involves implicit and…
Together AI built the world’s fastest speech-to-text stack.Parakeet on Together transcribes ~302 seconds of audio per second of processing time, the top speed factor reported by…
"Talk short. Drop grammar. Save token." This caveman style is widely promoted as a way to cut inference cost, but whether it actually saves anything depends…
As LLM agents increasingly select tools autonomously, their choices among tools with different privileges become safety-relevant. However, prior tool-selection studies focus on safety-agnostic metadata preferences, leaving…
Article URL: https://hl2.slqnt.dev/ Comments URL: https://news.ycombinator.com/item?id=48669534 Points: 78 # Comments: 17
OpenAI’s financial trajectory hinges heavily on infrastructure costs, a reality that drove the development of the new custom OpenAI Jalapeño chip. Developed in collaboration with Broadcom,…
Generating a coherent multi-shot video requires structured cross-shot memory. Subject appearance, scene context, and speaker identity must persist across cuts. Existing approaches either train end-to-end over…
Fine-grained visual reasoning requires multimodal large language models (MLLMs) to identify task-relevant visual evidence and ground their reasoning in local image regions. Existing agentic methods typically…
RT Arthur MacWatersHonestly at this point I'm a bit frustrated I didn't get my USAID-sponsored RPGArthur MacWaters: Naming it USAID was brilliant honestly because most people…
State of the art in AI image generation less than 5 years ago.Ethan Mollick: This is a pretty terrific computational image generator for making AI-created trading…
RT @levelsio: The craziest thing about Quake is how many modern FPS games originate from it somehowValve's Source engine was a modified Q…
Does X search just not work at all anymore? It used to sort-of work but now it doesn’t. Going to Grok for every search, even simple…
**Z.ai's GLM-5.2** leads in coding and agent benchmarks with top scores like **1595** on Code Arena: Frontend and **34.29%** reasoning accuracy with zero failures. Databricks improved…
Baidu open-sourced Unlimited OCR, a 3B-parameter MoE model that parses dozens of document pages in a single forward pass. Its Reference Sliding Window Attention (R-SWA) holds…
Great interviewJack Posobiec: BREAKING: DIRECTOR OF CITIZEN VIGILANTE JOINS HUMAN EVENTS DAILY EXCLUSIVE INTERVIEW https://x.com/i/broadcasts/1dRKZalmwzbxB
Thanks to Joshua Joseph, Dillon Plunkett, and Julian Huang for their feedback and for helping me refine these ideas.Anthropic recently reported that language models can “introspect.”…
RT X FreezeHow it started vs how it’s going
Wowstevenmarkryan: Bro took the red pill and I am here for it.Meet Karl Stefanovic.Three decades on Australian TV, including nearly 20 years co-hosting Nine’s flagship breakfast…
How come things like ROCm and the intel stack aren’t able to rapidly improve their software ecosystems to be a match for CUDA? Until the software…
RT LouI spend 10 minutes a day trying a new app. http://matrix.build is my favorite app today. You can tell from a product whether the team's…
RT Brooks OtterlakeI think part of what makes this account’s work so satisfying is they are doing a form of pattern recognition that is computerlike in…