FLOW EDIT
Easy editing of images with flow based models.
Full-text search across article titles and summaries. Refine with sector, year, content type, and reading-time filters.
Easy editing of images with flow based models.
Lucidrains has written up a great reimplementation of Meta's token + diffusion model Transfusion which can do images and text in a single model.
The article outlines the process of building an LLM inference engine using C++ and CUDA without external libraries, focusing on optimizing inference speed for consumer devices. It explores techniques like multithreading, vectorization, warp reductions, coalescing, and quantization to maximize performance and successfully surpass llama.cpp's throughput in specific scenarios. Additionally, the piece discusses potential for further optimization and the advantages of using established libraries for production-grade implementations.
Shifting from a model-as-person to a model-as-computer metaphor could enhance AI usefulness by enabling graphical interfaces and direct manipulation, rather than relying on slow conversational inputs. This new interaction paradigm could allow users to engage with AI like a dynamic, customizable app, offering more efficient and versatile functionality. Generative interfaces could eventually transform computing, allowing users to modify and create applications on demand for specific tasks.
This guide offers key insights from experts to help beginners get started with NotebookLM, making it easier to navigate and use effectively.
Anthropic's AI chatbot Claude is gaining popularity among tech insiders for its perceived emotional intelligence and creative responses.
Human-computer interaction is evolving from graphical interfaces to a more conversational AI-driven approach.
YouTube is introducing a feature for creators to permit third-party companies to train AI models using their videos.
Microsoft and OpenAI's definition of artificial general intelligence (AGI) is 'a system that can generate at least $100 billion in profits'. The companies agreed on the definition in 2023 as part of an agreement which had a clause that stated that OpenAI would stop allowing Microsoft to use any new technology it develops after AGI is achieved. OpenAI and Microsoft have been on diverging paths for some time now. Microsoft needs its own proprietary tech and OpenAI needs more money.
DeepSeek, a Chinese AI research lab, has released DeepSeek-V3, a Mixture-of-Experts model that features a total of 671B parameters. The model was trained on 14.8 trillion tokens. It has been released on GitHub along with a detailed paper outlining its capabilities. The model outperformed Meta's flagship Llama 3.1 405B parameter model and many other closed-source models on benchmark scores.
China launched 10 internet satellites on a Long March 5B rocket earlier this month. The satellites entered their predetermined orbits smoothly and the mission was a complete success. They are the first batch for Guowang, a broadband megaconstellation in low earth orbit (LEO) that will eventually consist of nearly 13,000 spacecraft. Guowang will be one of two Chinese versions of SpaceX's Starlink LEO megaconstellation.
Captura is a company that plans to electrochemically strip carbon dioxide out of the ocean, store or use it, then return the water to the sea, where it will naturally absorb more carbon dioxide from the air. It will start pumping as much of the Pacific ocean as it can in a few months at Hawaii's Keāhole Point. Besides Captura, there are dozens of similar field trials and pilot projects that will start scaling up their facilities in 2025.
Ghostty is a terminal emulator that doesn't force you to choose between speed, features, or native UIs. It can be used as a drop-in replacement for your existing terminal emulator. In addition to being a standalone terminal emulator, Ghostty is also a C-compatible library for embedding a fast, feature-rich terminal emulator in any 3rd party project.
awrit is a full graphical web browser for Kitty terminal with mouse and keyboard support. It works best with Kitty v0.28 or newer. awrit can be used to display documentation from DevDocs, watch the changes from Vite come to life, create a tiled layout without a tiling window manager, and add fancy UI using web technologies.
The release of o3 suggests the next few years are going to be insane. This post details some predictions about how AI will change the world over the next few years. The predictions cover areas such as AI agents, the software engineering industry, science and research, war and societal chaos, what happens if AI goes rogue, and more.
Google may have bought Character AI because Character AI had figured out how to run its models without costing a fortune. Character AI was able to serve millions of users without bleeding money. Google is now giving away flash models for free on AI Studio - this wouldn't be possible without figuring out how to make inference really cheap. Pre-training has hit diminishing returns, so companies are shifting toward chain-of-thought prompting and inference-time optimizations.
ChatGPT search is surprisingly competitive when it comes to informational queries, but Google is better at almost all queries except content gap analysis.
While Apple keeps adding cool new software features, there still isn't an immediate, obvious reason to buy a Vision Pro.
Loyal, a US biotech start-up, plans to release LOY-002, a pill that could give dogs a minimum of one extra year of healthy life, early next year.
POSIX time assumes the length of the day is 86,400 seconds, which is not always true.
DeepSeek-V3's training only cost $5.576 million - it took around 55 days on a 2048xH800 cluster.
The optimal approach for a domain when considering both performance and operational resilience is for the domain to be served by two distinct dual-stack diverse anycast nameservers.
Google contractors are comparing Gemini AI's responses with Anthropic's Claude, but it's unclear if Google has permission to use Claude for testing. The contractors noted that Claude prioritizes safety in its responses more than Gemini. Google states that while it evaluates model outputs against competitors, it does not train Gemini on Anthropic's models.
AI systems are rapidly advancing, outperforming expectations on new challenging evaluations, like Epoch AI's FrontierMath. Despite this progress, designing effective evals to understand and manage AI capabilities remains complex and underfunded. Experts stress the need for sophisticated, timely tests to monitor potential risks as models evolve.
AI hallucinations, often criticized for inaccuracies, are proving valuable in scientific research by accelerating idea generation and discovery processes. Notable achievements include Nobel Prize-winning protein designs and innovations in antibiotic development and catheter design. Despite controversies around the term "hallucinations," experts acknowledge AI's potential for groundbreaking scientific insights.
Meta has released an updated agents framework to measure and ensure robustness and safety when deployed in the wild.
This research introduces a technique for applying material transformations, like wetness or coating, across different scenes using disentangled Neural Radiance Fields (NeRF).
Researchers explored the consequences of training AI language models to optimize for user feedback, like thumbs up ratings. The study shows that this approach can lead to manipulation as AIs learn to game the system.
A new benchmark evaluates the federated fine-tuning of MLLMs across diverse scenarios, including two datasets, five baselines, and over ten types of multimodal heterogeneities.
A helpful guide for prompt engineering.
A number of puzzles and tutorials to learn GPU programming on Mac Metal acceleration.
Fei-Fei Li's startup, World Labs, aims to enhance AI with 3D spatial intelligence to generate and interact with 3D worlds. The focus on spatial intelligence is crucial for advancing AI capabilities in both real and virtual 3D environments. This technology could revolutionize sectors like robotics, design, and augmented reality.
AI development faces a data shortage crisis, projected to hit by 2028, as training data sizes approach the total stock of public online text. AI companies like OpenAI are exploring solutions such as generating new data and using unconventional sources to mitigate this issue. This data crunch may shift focus toward smaller, specialized AI models rather than large-scale LLMs.
Red Rabbit Robotics is advancing the field of autonomous labor by developing and open-sourcing the RX1 robot for manufacturing and commercial applications. It aims to tap into the labor shortage by providing cost-effective, robotic solutions for dull, dangerous, and dirty jobs. The company's goal is to make robots more accessible at a lower price point and gradually transition from teleoperation to full autonomy, prioritizing utility and widespread adoption.
Klarna CEO Sebastian Siemiatkowski claims generative AI has enabled a workforce reduction, though the company is still hiring for essential roles.
Researchers at Meta and UC San Diego are developing LLMs that process logical solutions in "latent space," bypassing natural language constraints.
OpenAI's Kubernetes incident on December 11 was caused by unexpected interactions, where a new telemetry service overloaded the Kubernetes API servers, leading to failures in DNS-based service discovery.
Amazon has reportedly instructed some of its Chinese cross-border merchants to stop listing lower-priced products on Temu. Temu has seen rapid growth in its market share in the US - it was the most downloaded free app in the country in 2024. The site's aggressive pricing strategy has forced Amazon to take action. Amazon has responded to the rumors by stating that sellers have the freedom to determine their own sales strategies and can set prices for their products on Amazon and other sales channels.
Meta's Ray-Ban smart glasses may get displays as soon as the second half of 2025. The new displays will be used to show notifications and responses from Meta's virtual assistant. Plans to turn Orion, its augmented reality glasses prototype, into a consumer product have been brought forward, though a release is still likely to be years away. More big tech firms, like Qualcomm, Samsung, and Google, are entering into the mixed-reality glasses space. Apple may be releasing a cheaper version of the Apple Vision Pro at some point after 2027.
Animal communications have patterns and structures that convey information to other members of that animal's species. A growing number of biologists and computer scientists are applying AI to animal sounds to reveal what they are saying to each other. AI-assisted studies have found that both African savannah elephants and common marmoset monkeys give names to their companions. As these computer models improve, they may enable scientists to investigate animals' self-awareness.
Chinese scientists have turned bamboo into a transparent material that is resistant to fire and water and suppresses smoke. The team used the same method used to turn wood transparent to turn the bamboo transparent - bamboo's internal structure and chemical composition is very similar to wood. The resulting transparent bamboo can be used as a building material. When used as a substrate for perovskite solar cells, it acts like a light management layer, boosting cells' power conversion efficiency by 15.29%.
Shortest is an AI-powered natural language end-to-end testing framework. It lets developers write tests in plain English and let AI handle the execution. Shortest uses Anthropic's Claude API. It is built on Playwright and has GitHub integration with 2FA support.
2024 was a great year for CSS. Several features, including cross-document view transitions, scroll-driven animations, anchor positioning, and animate to height: auto were introduced. This article suggests more CSS features that developers want, such as a conditional if() statement, CSS mixins, anchor positioning in more browsers, and a way to balance flex-wrap.
Evaluations (or evals) help researchers understand what advanced AI systems can do. Designing effective evals is challenging and expensive and the field is underfunded. Modern AI systems regularly achieve top scores on many popular evals, making it harder to judge how quickly they are improving. Companies, nonprofits, and governments have created new sets of much more challenging evals in response, but AI systems are making astonishing progress, even on the most advanced tests.
Meta said in April that it would raise spending levels by as much as $10 billion to support infrastructure investments for its AI efforts. The announcement sent shares plunging, but several months later, Meta's stock price has hit a new record. AI has become Meta's top priority, directly impacting the company's business and potentially paving the way for future revenue opportunities. This all-encompassing approach to AI has led analysts to predict that Meta will be positioned for more success in 2025.
Investors are enthusiastic about artificial intelligence as they predict that it will result in a supercycle of iPhone upgrades.
A thread full of links to resources for learning general purpose GPU development on Apple's M-series chips.
Server-Sent Events, a one-way communication channel from server to client over HTTP, offer a simpler, often overlooked alternative to WebSockets.
This article provides a history of antibiotics, describes how they were developed, explains why antibiotic development has slowed down, and gives some ideas on how to reignite antibiotic development today.
GIF virtually vanished from the Internet for about 10 years due to it being patented.