Stripe will reportedly acquire AI gateway startup OpenRouter for $7B+
OpenRouter's CEO recently described the startup as Stripe for AI.
Full feed
Feed 25,494 posts
Every story across every category, newest first. Each card links to the original publisher; daily-brief posts open as editorial pages.
No, DeepSeek definitely didn’t misprice cache reads. They are a research lab, and every release is a research release. This is Liang making use of th…
No, DeepSeek definitely didn’t misprice cache reads.They are a research lab, and every release is a research release. This is Liang making use of the terms…
NInfer RTX 4090 for Qwen 3.8 27B update – up to 250-350K tokens context in VRAM
I've made some improvements to my fork of NInfer, adding rk2v4-e8 quant option for the KV cache, which can reach up to 250-350K tokens of context…
Fascinating
FascinatingTeortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞): One of the more underrated thinkers of the ageUnderrated even by me as I say this (I haven’t read most…
The Case Against Formal Verification, 50 Years Later
Article URL: https://ivan-gavran.github.io/0-social-processes-paper Comments URL: https://news.ycombinator.com/item?id=49323459 Points: 6 # Comments: 1
RT Ravid Shwartz Ziv: There are two separate questions regarding Dario Amodei’s post about AI and biology: motivation and reality. Motivation. Dario …
RT Ravid Shwartz ZivThere are two separate questions regarding Dario Amodei’s post about AI and biology: motivation and reality.Motivation. Dario started in biology (was a PhD…
Why people aren’t buying Mark Zuckerberg’s AI future
On the latest episode of Equity podcast, we discuss why not everyone is buying Zuckerberg’s vision.
Stripe Clinches over $7B Deal to Buy AI Firm OpenRouter
Article URL: https://www.bloomberg.com/news/articles/2026-08-16/stripe-nears-deal-to-buy-ai-firm-openrouter-for-over-7-billion Comments URL: https://news.ycombinator.com/item?id=49323381 Points: 14 # Comments: 6
An open-source, candle-only crypto pump detector–looking for ML contributors
Article URL: https://github.com/aleks-ent/pump-dump-crypto-screener Comments URL: https://news.ycombinator.com/item?id=49323256 Points: 3 # Comments: 0
RT Neil Stone: "There are no Amish with autism" There are "Vaccines aren't tested against placebo" They are "MMR has never been studied as a possible …
RT Neil Stone"There are no Amish with autism"There are "Vaccines aren't tested against placebo"They are "MMR has never been studied as a possible cause of autism"It…
ICE Shot a Journalist and Threw Him in Detention. He's Approaching 300 Days
Article URL: https://theintercept.com/2026/08/16/ricardo-parias-ice-detention-journalist-los-angeles/ Comments URL: https://news.ycombinator.com/item?id=49323157 Points: 27 # Comments: 1
Red queen hypothesis – a new way forward for self-improving AI
Article URL: https://www.cst.cam.ac.uk/news/red-queen-hypothesis-new-way-forward-self-improving-ai Comments URL: https://news.ycombinator.com/item?id=49323136 Points: 11 # Comments: 1
They are matching prices
They are matching pricesTogether AI: DeepSeek V4 Pro 0813 is live on Together AIDeepSeek’s flagship V4 Pro release brings a 1.6T MoE architecture, 1M context, and…
The Trumps' Crypto Project Just Got One Step Closer to Becoming a Bank
Article URL: https://www.motherjones.com/politics/2026/08/donald-trump-world-liberty-regulatory-approval/ Comments URL: https://news.ycombinator.com/item?id=49322979 Points: 14 # Comments: 12
Qwen3.8-27b on RTX 3090 – 82 tps single request, up to 672 tps peak
Hi, After a long night of optimizations, I believe I have made the fastest inference engine for Qwen3.6-28B on a 3090. Quick metrics: - 250w power…
Why are RTX 6000 PROs still getting bought at 16000+ USD? And who are buying them?
Hello guys, hoping you're doing well. I bring this discussion since I have noticed on internet, be USA or EU, RTX 6000 PROs at 16000USD or…
Qwen3.8 27B Q2 vs Q3 vs Qwen3.6 35B-A3B MoE on 12GB VRAM
Did a quick local test because I wanted to see what is actually usable on my 12GB laptop GPU. I tested the newer Qwen3.8 27B dense…
Qwen 3.8-27b unusable long thinking?
I have a small coding test, where I ask a model to implement a simple CLI from a spec file. Qwen3.6-27b can do it in ~50k…
Models Are Getting Dumber on Purpose
Article URL: https://w4g1.dev/blog/models-are-getting-dumber-on-purpose Comments URL: https://news.ycombinator.com/item?id=49322695 Points: 59 # Comments: 19
Q2.5 2026 Timelines Update: Uplift and Revenue
Tl;dr: Our timelines haven’t changed much (they got slightly shorter) but our modeling and evidence base have noticeably improved, so we feel somewhat more confident.SummaryWe intend…
IMO OLMo 1-3 still underrated in their contributions to science. I wish we could’ve spent more time amplifying this while building them.
IMO OLMo 1-3 still underrated in their contributions to science. I wish we could’ve spent more time amplifying this while building them.
Protobuf has LSP support. You're welcome
Article URL: https://buf.build/blog/protobuf-lsp Comments URL: https://news.ycombinator.com/item?id=49322573 Points: 22 # Comments: 4
Case for Funding AI Safety in Japan
Intro and tl;drI'm Esa Koskinen, working as a volunteer director of AI Safety Tokyo.Epistemic status: estimates by an interested party; I run one of the organizations…
Will There Be an AI Hegemon? A Mental Model for AI Power Concentration
On the Baker-Anthropic Conversation, Power Acquisition and Escape VelocityWill there be an AI hegemon, i.e. an entity that, by wielding superintelligence, acquires so much economic or…