GPTZero finds 100 new hallucinations in NeurIPS 2025 accepted papers

Why does SSH send 100 packets per keystroke?

Turso is an in-process SQL database, compatible with SQLite

TI-99/4A: Leaning More on the Firmware

Improving the usability of C libraries in Swift

Why medieval city-builder video games are historically inaccurate (2020)

CSS Optical Illusions

Writeup (includes good&#x2F;bad sample generations): <a href="https:&#x2F;&#x2F;www.linum.ai&#x2F;field-notes&#x2F;launch-linum-v2">https:&#x2F;&#x2F;www.linum.ai&#x2F;field-notes&#x2F;launch-linum-v2</a>We&#x27;re Sahil and Manu, two brothers who spent the last 2 years training text-to-video models from scratch. Today we&#x27;re releasing them under Apache 2.0.These are 2B param models capable of generating 2-5 seconds of footage at either 360p or 720p. In terms of model size, the closest comparison is Alibaba&#x27;s Wan 2.1 1.3B. From our testing, we get significantly better motion capture and aesthetics.We&#x27;re not claiming to have reached the frontier. For us, this is a stepping stone towards SOTA - proof we can train these models end-to-end ourselves.Why train a model from scratch?We shipped our first model in January 2024 (pre-Sora) as a 180p, 1-second GIF bot, bootstrapped off Stable Diffusion XL. Image VAEs don&#x27;t understand temporal coherence, and without the original training data, you can&#x27;t smoothly transition between image and video distributions. At some point you&#x27;re better off starting over.For v2, we use T5 for text encoding, Wan 2.1 VAE for compression, and a DiT-variant backbone trained with flow matching. We built our own temporal VAE but Wan&#x27;s was smaller with equivalent performance, so we used it to save on embedding costs. (We&#x27;ll open-source our VAE shortly.)The bulk of development time went into building curation pipelines that actually work (e.g., hand-labeling aesthetic properties and fine-tuning VLMs to filter at scale).What works: Cartoon&#x2F;animated styles, food and nature scenes, simple character motion. What doesn&#x27;t: Complex physics, fast motion (e.g., gymnastics, dancing), consistent text.Why build this when Veo&#x2F;Sora exist? Products are extensions of the underlying model&#x27;s capabilities. If users want a feature the model doesn&#x27;t support (character consistency, camera controls, editing, style mapping, etc.), you&#x27;re stuck. To build the product we want, we need to update the model itself. That means owning the development process. It&#x27;s a bet that will take time (and a lot of GPU compute) to pay off, but we think it&#x27;s the right one.What’s next?
- Post-training for physics&#x2F;deformations
- Distillation for speed
- Audio capabilities
- Model scalingWe kept a “lab notebook” of all our experiments in Notion. Happy to answer questions about building a model from 0 → 1. Comments and feedback welcome!

Show HN: Text-to-video model from scratch (2 brothers, 2 years, 2B params)

Stunnel

'Active' sitting is better for brain health: review of studies

In Europe, wind and solar overtake fossil fuels

Hey HN! We&#x27;re Nithin and Nikhil, twin brothers building BrowserOS (YC S24). We&#x27;re an open-source, privacy-first alternative to the AI browsers from big labs.The big differentiator: on BrowserOS you can use local LLMs or BYOK and run the agent entirely on the client side, so your company&#x2F;sensitive data stays on your machine!Today we&#x27;re launching filesystem access... just like Claude Cowork, our browser agent can read files, write files, run shell commands! But honestly, we didn&#x27;t plan for this. It turns out the privacy decision we made 9 months ago accidentally positioned us for this moment.The architectural bet we made 9 months ago: Unlike other AI browsers (ChatGPT Atlas, Perplexity Comet) where the agent loop runs server-side, we decided early on to run our agent entirely on your machine (client side).But building everything on the client side wasn&#x27;t smooth. We initially built our agent loop inside a Chrome extension. But we kept hitting walls -- service worker being single thread JS; not having access to NodeJS libraries. So we made the hard decision 2 months ago to throw away everything and start from scratch.In the new architecture, our agent loop sits in a standalone binary that we ship alongside our Chromium. And we use gemini-cli for the agent loop with some tweaks! We wrote a neat adapter to translate between Gemini format and Vercel AI SDK format. You can look at our entire codebase here: <a href="https:&#x2F;&#x2F;git.new&#x2F;browseros-agent" rel="nofollow">https:&#x2F;&#x2F;git.new&#x2F;browseros-agent</a>How we give browser access to filesystem: When Claude Cowork launched, we realized something: because Atlas and Comet run their agent loop server-side, there&#x27;s no good way for their agent to access your files without uploading them to the server first. But our agent was already local. Adding filesystem access meant just... opening the door (with your permissions ofc). Our agent can now read and write files just like Claude Code.What you can actually do today:a) Organize files in my desktop folder <a href="https:&#x2F;&#x2F;youtu.be&#x2F;NOZ7xjto6Uc" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;NOZ7xjto6Uc</a>b) Open top 5 HN links, extract the details and write summary into a HTML file <a href="https:&#x2F;&#x2F;youtu.be&#x2F;uXvqs_TCmMQ" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;uXvqs_TCmMQ</a>--- Where we are now
If you haven&#x27;t tried us since the last Show HN (<a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=44523409">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=44523409</a>), give us another shot. The new architecture unlocked a ton of new features, and we&#x27;ve grown to 8.5K GitHub stars and 100K+ downloads:c) You can now build more reliable workflows using n8n-like graph <a href="https:&#x2F;&#x2F;youtu.be&#x2F;H_bFfWIevSY" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;H_bFfWIevSY</a>d) You can also use BrowserOS as an MCP server in Cursor or Claude Code <a href="https:&#x2F;&#x2F;youtu.be&#x2F;5nevh00lckM" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;5nevh00lckM</a>We are very bullish on browser being the right platform for a Claude Cowork like agent. Browser is the most commonly used app by knowledge workers (emails, docs, spreadsheets, research, etc). And even Anthropic recognizes this -- for Claude Cowork, they have janky integration with browser via a chrome extension. But owning the entire stack allows us to build differentiated features that wouldn&#x27;t be possible otherwise. Ex: Browser ACLs.Agents can do dumb or destructive things, so we&#x27;re adding browser-level guardrails (think IAM for agents): &quot;role(agent): can never click buy&quot; or &quot;role(agent): read-only access on my bank&#x27;s homepage.&quot;Curious to hear your take on this and the overall thesis.We’ll be in the comments. Thanks for reading!GitHub: <a href="https:&#x2F;&#x2F;github.com&#x2F;browseros-ai&#x2F;BrowserOS" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;browseros-ai&#x2F;BrowserOS</a>Download: <a href="https:&#x2F;&#x2F;browseros.com">https:&#x2F;&#x2F;browseros.com</a> (available for Mac, Windows, Linux!)

Show HN: BrowserOS – "Claude Cowork" in the browser

Design Thinking Books (2024)

AnswerThis (YC F25) Is Hiring

Keeping 20k GPUs healthy

Anthropic Economic Index economic primitives

The Cscript Style Guide – CScript is the standard C

The Uncomfortable Math of Working for Yourself

Hey HN! This has been my project for a few years now. I recently brought it back to life after taking a pause to focus on my studies.My goal with this project is to separate fluff from science when shopping for supplements. I am doing this in 3 steps:1.) I index every supplement on the market (extract each ingredient, normalize by quantity)2.) I index every research paper on supplementation (rank 
every claim by effect type and effect size)3.) I link data between supplements and research papersEarlier last year, I took pause on a project because I&#x27;ve ran into a few issues:Legal: Shady companies are sending C&amp;Ds letters demanding their products are taken down from the website. It is not something I had the mental capacity to respond to while also going through my studies. Not coincidentally, these are usually brands with big marketing budgets and poor ingredients to price ratio.Technical: I started this project when the first LLMs came out. I&#x27;ve built extensive internal evals to understand how LLMs are performing. The hallucinations at the time were simply too frequent to passthrough this data to visitors. However, I recently re-ran my evals with Opus 4.5 and was very impressed. I am running out of scenarios that I can think&#x2F;find where LLMs are bad at interpreting data.Business: I still haven&#x27;t figured out how to monetize it or even who the target customer is.Despite these challenges, I decided to restart my journey.My mission is to bring transparency (science and price) to the supplement market. My goal is NOT to increase the use of supplements, but rather to help consumers make informed decisions. Often times, supplementation is not necessary or there are natural ways to supplement (that&#x27;s my focus this quarter – better education about natural supplementation).Some things that are helping my cause – Bryan Johnson&#x27;s journey has drawn a lot more attention to healthy supplementation (blueprint). Thanks to Bryan&#x27;s efforts, I had so many people in recent months reach out to ask about the state of the project – interest I&#x27;ve not had before.I am excited to restart this journey and to share it with HN. Your comments on how to approach this would be massively appreciated.Some key areas of the website:* Example of navigating supplements by ingredient <a href="https:&#x2F;&#x2F;pillser.com&#x2F;search?q=%22Vitamin+D%22&amp;s=jho4espsuc" rel="nofollow">https:&#x2F;&#x2F;pillser.com&#x2F;search?q=%22Vitamin+D%22&amp;s=jho4espsuc</a>* Example of research paper analyzed using AI <a href="https:&#x2F;&#x2F;pillser.com&#x2F;research-papers&#x2F;effect-of-lactobacillus-gasseri-pa-168-bifidobacterium-longum-sp-073-b-bifidum-mf-205-on-common-cold-episodes-a-double-blind-randomized-controlled-trial-767" rel="nofollow">https:&#x2F;&#x2F;pillser.com&#x2F;research-papers&#x2F;effect-of-lactobacillus-...</a>* Example of looking for very specific strains or ingredients <a href="https:&#x2F;&#x2F;pillser.com&#x2F;probiotics&#x2F;bifidobacterium-bifidum" rel="nofollow">https:&#x2F;&#x2F;pillser.com&#x2F;probiotics&#x2F;bifidobacterium-bifidum</a>* Example of navigating research by health-outcomes <a href="https:&#x2F;&#x2F;pillser.com&#x2F;health-outcomes&#x2F;improved-intestinal-barrier-function" rel="nofollow">https:&#x2F;&#x2F;pillser.com&#x2F;health-outcomes&#x2F;improved-intestinal-barr...</a>* Example of product listing <a href="https:&#x2F;&#x2F;pillser.com&#x2F;supplements&#x2F;pb-8-probiotic-663" rel="nofollow">https:&#x2F;&#x2F;pillser.com&#x2F;supplements&#x2F;pb-8-probiotic-663</a>

Show HN: I've been using AI to analyze every supplement on the market

Doctors in Brazil using tilapia fish skin to treat burn victims (2017)

6:15:29 AM FRIDAY JANUARY 23, 2026

THE HACKER NEWS