Try Grok Build http://X.ai/build
Try Grok Buildhttp://X.ai/buildX Freeze: Grok Build officially reaches v1.0.0, rapidly becoming one of the most powerful coding harnesses in the worldAfter 100+ updates in just over…
Latest
DeepSeek V4 Flash 0728 really is the current cost-performance frontier!
DeepSeek V4 Flash 0728 really is the current cost-performance frontier!Hassan: DeepSeek V4 Flash is 6x cheaper than GPT 5.6 Luna for coding tasks!Luna scores higher on…
Airbnb says AI is helping it ship features faster as it tests a new search function
Airbnb will debut a new AI-powered search experience with a toggle.
Agenda: Infrastructure for Trading with Partially Misaligned AIs
This post can be read on its own, without checking the rest of this sequence.Unlike the other parts of the sequence, this post is mostly not…
Wan-Animate-2: Pushing the Application Boundaries of Character Animation Models
📝 Introduction We present Wan-Animate-2, a novel end-to-end character animation framework that directly consumes driving videos in a redesigned Diffusion Transformer, which achieves high-fidelity motion generation…
b10308
Mitigate crashing issue on Windows MSYS2 UCRT64 environment (GCC 16.1.0) (#26555) Website: https://llama.app macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS…
US's largest newspaper chain partners with Palantir to analyze audience data
Article URL: https://www.niemanlab.org/2026/08/americas-largest-newspaper-chain-usa-today-co-partners-with-palantir-to-analyze-audience-data/ Comments URL: https://news.ycombinator.com/item?id=49210589 Points: 44 # Comments: 10
alphaXiv is the future, have been a power user for a while
alphaXiv is the future, have been a power user for a whileAlbert Jiang: https://www.alphaxiv.org/abs/2608.leanstralLeanstral paper now on alphaXiv where we present training and evals in their…
Jill Lepore on the ‘Artificial State’ and why Silicon Valley’s leaders are bad sci-fi readers
Historian Jill Lepore has a theory about why tech companies often use soaring language to describe their products — almost as if they’re forming a new…
nice rl experiment on train-inference mismatch
nice rl experiment on train-inference mismatchYichuan Wang: Zero Train–Inference Mismatch — now for linear attention, and under async RL 🎯We got bitwise-exact trainer/generator parity for Gated…
AI chatbots have failed people in crisis. Can that be fixed?
Clinicians and researchers say AI companies need to open up their safety data.
Noticing more and more of this Young foreign born Chinese go turbo Chinamaxxing. De-assimilation. This one even became ESL and lost the capability to …
Noticing more and more of thisYoung foreign born Chinese go turbo Chinamaxxing. De-assimilation. This one even became ESL and lost the capability to parse sarcasm. Didn’t…
U.S. Lost 23,000 Jobs in July, While Unemployment Ticked Lower
Article URL: https://www.wsj.com/economy/jobs/july-jobs-report-unemployment-0d7c08a7 Comments URL: https://news.ycombinator.com/item?id=49210240 Points: 36 # Comments: 1
RT Emma Scharfman: A new foundation model that reads and generates DNA sequences was just released on the Hugging Face: https://huggingface.co/marin-d…
RT Emma ScharfmanA new foundation model that reads and generates DNA sequences was just released on the Hugging Face: https://huggingface.co/marin-dna/marin-dna-scaling-v0.5-h1920-p1B Feel free to check the test…
Uhh… hello, Based Department?
Uhh… hello, Based Department?Elon Musk: @beffjezos FEL FTW
RT Steve Hou: The fact that China is increasingly allowing the narrative to surface is itself instructive. @DavidDaokuiLi is pointing out what many of…
RT Steve HouThe fact that China is increasingly allowing the narrative to surface is itself instructive. @DavidDaokuiLi is pointing out what many of us have known…
ByteDance trains massive AI model in bid to rival Anthropic
TikTok owner training a model with 10 trillion parameters.
AI psychosis is the new leadership blind spot
Article URL: https://www.fastcompany.com/91576086/ai-psychosis-is-the-new-leadership-blind-spot-ai-leadership-blind-spots Comments URL: https://news.ycombinator.com/item?id=49210077 Points: 19 # Comments: 3
Upcoming Dovetail fellow talks & discussion
As the current Dovetail research fellowship comes to a close, the fellows are giving talks on their projects. All are welcome to join!The easiest way to…
RT Gabriel Peyré: The Mathematical Nexus brings together 800 animated vignettes and 140 accompanying Python notebooks, encompassing most of the mathe…
RT Gabriel PeyréThe Mathematical Nexus brings together 800 animated vignettes and 140 accompanying Python notebooks, encompassing most of the mathematical content I have shared on social…
b10307
sycl: fix UE4M3 parsing (#25608) The NVFP4 quantization format stores a scaling factor for every group of 16 weights, packed into a single UE4M3 byte. The…
LFM2.5-2.6B model+KV cache quantization report
LFM2.5-2.6B is a new tiny model by LiquidAI, with benchmarks that put it head to head with much larger models. I've run llama-perplexity on many model…
The web server deployment model breaks at hobby scale
Article URL: https://w.on-t.work/web-deployment-model Comments URL: https://news.ycombinator.com/item?id=49209921 Points: 22 # Comments: 12
Built a tool to generate slides from research papers using local LLMs (because I hate formatting decks and privacy matters) [P]
Hi guys, Every time I had to prepare a presentation based on a paper or research doc, I found the process super tedious. Plus, I really…