2371
EleutherAI announces it has become a non-profit
(Some of) what we've been up to for the past year-and-a-half at EleutherAI.
Top posts
Every blog post that took off on Hacker News, ranked by the votes of its best submission. 6,328 posts.
2372
2373
2374
Chat Notebooks
Wolfram expands Notebooks, integrating LLM functionality into the new chat cell and Chat Notebook. Stephen Wolfram explains how chat-enabled and chat-driven versions work.
2375
Tensorflow on edge, or – Building a “smart” security camera with a Raspberry Pi
The amount of time my outdoor cameras are being set off by light, wind, cars, or anything other than a human is insane. Overly cautious security cameras might be a feature, but an annoying one at that...
2376
What we learned in Seoul with AlphaGo
DeepMind's AlphaGo took on and defeated legendary Go player, Lee Sedol, marking a major milestone for artificial intelligence.
2377
Neuron, a new VS Code extension for data science
Access to this page requires authorization. You can try signing in or changing directories .
2378
Transit rewards
Today, Waymo is excited to announce our transit rewards program, which we will initially offer in the San Francisco Bay Area, with future cities to follow. This first-of-its-kind initiative rewards riders with Waymo Cash when they connect their Waymo rides with public transit and use their Visa card.
2379
Using AI to reverse engineer hardware
After my mother passed away I ended up with a big box full of old film negatives containing photos from childhood family trips and similar. So I decided to buy a negative photo scanner and found that the Plustek 8200i was supported under Linux so I bought that. Only to discover that the chipset in...
2380
Migrating a production AI agent to GPT-5.6: 2.2x faster, 27% cheaper
For four months, no frontier model beat Claude Opus in our production evals. GPT-5.6 did. This is what we learned while migrating.
2381
FLUX is fast and it's open source
FLUX is now much faster on Replicate, and we’ve made our optimizations open-source so you can see exactly how they work and build upon them.
2382
Benchmarking TensorFlow on Cloud CPUs: Cheaper Deep Learning Than Cloud GPUs
Using CPUs instead of GPUs for deep learning training in the cloud is cheaper because of the massive cost differential afforded by preemptible instances.
2383
Mercury 3: Diffusion at scale
Mercury 3 is the most capable diffusion LLM released to date. It outperforms GPT-6 Luna and Claude Haiku 5.5 while running 4–8x faster at 1,000+ tokens per second.
2384
2385
How has DeepSeek improved the Transformer architecture?
This Gradient Updates issue goes over the major changes that went into DeepSeek's most recent model.
2386
2387
NHL goal celebration hack: real-time machine learning and hue light show
This post does a deep dive on the **epic goal celebration hack**. In a single sentence, I trained a machine learning model to detect in real-time that a goal was just scored by the Habs based on the live audio feed of a game and to trigger a light show using Philips hues in my living room.
2388
Takeaways from hundreds of LLM finetuning experiments with LoRA
From the PyTorch Lightning creators. Own your AI, don't rent it. Lightning is the AI cloud to build, train, and deploy your own models on 36,000+ NVIDIA GPUs.
2389
2390
The Engineer’s Guide to Deep Learning: Understanding the Transformer Model
In the previous two golden ages (1950s-1960s and the 1980s), our expectations outpaced the capabilities of the technology at the time, leading to disappointment. In contrast, the AI technology of the current golden age, which began in the mid-2010s, has consistently exceeded our expectations.
2391
Testing how hard it is to cheat with ChatGPT in interviews
We had interviewees use ChatGPT without their interviewer knowing. Here's what happened.
2392
Google, Nvidia, and OpenAI
OpenAI and Nvidia are both under threat from Google; I like OpenAI’s chances best, but they need an advertising model to beat Google as an Aggregator.
2393
The Pentagon is making a mistake by threatening Anthropic
Anthropic faces a Friday deadline to allow domestic surveillance and automated killer robots.
2394
CriticGPT: Finding GPT-4's mistakes with GPT-4
CriticGPT, a model based on GPT-4, writes critiques of ChatGPT responses to help human trainers spot mistakes during RLHF
2395
Giving university exams in the age of chatbots
Giving University Exams in the Age of Chatbots par Ploum - Lionel Dricot.
2396
MUM: A new AI milestone for understanding information
Our latest AI milestone, Multitask Unified Model, has the potential to take Google’s information understanding to a new level.
2397
Agents for financial services and insurance
Ten new Cowork and Claude Code plugins, Microsoft 365 integrations, new connectors, and an MCP app for financial services and insurance.
2398
How we made our AI code review bot stop leaving nitpicky comments
Lessons learned from our challenges reducing LLM output: Discover everything that went wrong when trying to make our models say less.
2399
How we built the Waifu Vending Machine
On the weekend of Anime Expo, an artist promised the humanly impossible -- custom art commissions with a five-minute turnaround time. The not-so-hidden catch? The artist wasn't a human at all, but an AI exposed to millions of anime images.
2400
Fable 5 vs. GPT-5.6 Sol on an NP-Hard Problem: Does /goal help?
I gave Claude Fable 5 and GPT-5.6 Sol the same unpublished NP-hard optimization problem, with and without their native /goal mode. Fable 5 is a beast; /goal is not a game changer.