hackernews

1. Claude Sonnet 4.5 Total comment counts : 61 Summary Claude Sonnet 4.5 is presented as the world’s best coding model and a top option for building complex agents and reasoning with computers. It debuts major upgrades: Claude Code adds checkpoints, refreshed terminal, a native VS Code extension, a new context-editing feature, and a memory tool; Claude apps enable code execution and file creation in-chat; and Claude for Chrome is available to Max users on the waitlist....

1. Privacy Badger is a free browser extension made by EFF to stop spying Total comment counts : 9 Summary Privacy Badger is an Electronic Frontier Foundation browser extension that blocks third-party trackers across sites, not general ads. It automatically blocks advertisers who track you across multiple sites. Unlike traditional blockers, it uses an algorithmic definition of tracking and learns from behavior if trackers ignore Do Not Track and Global Privacy Control signals....

1. A WebGL game where you deliver messages on a tiny planet Total comment counts : 49 Summary error Overall Comments Summary Main point: A browser-based game with stunning visuals and atmosphere that emphasizes exploration and nonverbal multiplayer delivery, attracting broad praise and curiosity. Concern: Concerns include clumsy controls and camera behavior, unclear delivery objectives, and a missing or weak tutorial that could hinder new players. Perspectives: Opinions range from awe at the art, performance, and potential, to practical critiques about controls, onboarding, and curiosity about the development tools and engine....

1. SimpleFold: Folding proteins is simpler than you think Total comment counts : 19 Summary SimpleFold is a flow-matching-based protein folding model that uses only general transformer layers, avoiding expensive modules like triangle attention or pair biases. Trained with a generative flow-matching objective, it scales to 3B parameters and learns from over 8.6 million distilled structures plus experimental PDB data, making it the largest-scale folding model to date. It achieves competitive performance on standard benchmarks and strong ensemble predictions, challenging the idea that domain-specific architectures are required....

1. Improved Gemini 2.5 Flash and Flash-Lite Total comment counts : 25 Summary Google announces updated Gemini 2.5 Flash and 2.5 Flash-Lite on AI Studio and Vertex AI, with testing previews gemini-2.5-flash-lite-preview-09-2025 and gemini-2.5-flash-preview-09-2025. A -latest alias now points to the newest releases to simplify access, with a mandatory 2-week notice before updates or deprecation behind -latest. Previews are for testing and feedback and are not guaranteed stable; for stability, use gemini-2....

1. Just let me select text Total comment counts : 106 Summary A lonely user on Bumble laments how non-selectable text in a profile makes reading a pretty German girl’s bio nearly impossible. He must screenshot and OCR to extract it, turning living language into opaque images. He argues that text is fundamental—copyable, translatable, accessible—and that disabling it degrades user comprehension and accessibility. Calling it a “crime against the user,” he urges designers to keep text selectable and invites feedback....

1. Baldur’s Gate 3 Steam Deck – Native Version Total comment counts : 11 Summary Hotfix #34 on Steam Deck installs Baldur’s Gate 3 Native. Proton runs via the Steam Deck Proton layer and uses more CPU; Native uses less. You can revert to Proton if needed. Steps: Steam page → Settings → Properties. Linux support: Larian does not provide Linux support; Native is Steam Deck‑only. Saves: before Native, in compatdata; after, at /home/deck/....

1. Qwen3-Omni: Native Omni AI model for text, image and video Total comment counts : 10 Summary Qwen3-Omni is Alibaba Cloud’s end-to-end multilingual omni-modal LLM that understands and generates text, audio, images, and video, with real-time streaming in text and speech. Built with MoE Thinker–Talker and AuT pretraining, it achieves state-of-the-art results across modalities and supports 119 text languages, 19 speech input languages, and 10 speech outputs. It enables low-latency multimodal interactions, customizable behavior via system prompts, and an open-source audio-captioning model (Qwen3-Omni-30B-A3B-Captioner)....

1. Sj.h: A tiny little JSON parsing library in ~150 lines of C99 Total comment counts : 16 Summary This article introduces a small, free JSON parsing library released into the public domain. It includes a tiny library and a program that loads a Rect struct from a JSON string, with demos in a folder. The project emphasizes user feedback and documentation, though the page occasionally shows loading errors. Overall Comments Summary Main idea: The discussion centers on small, single-file libraries in ANSI C or Lua that have focused scope, good docs, and free software licenses....

1. Ultrasonic Chef’s Knife Total comment counts : 31 Summary Seattle Ultrasonics is promoting a Batch 1 pre-order with estimated shipping by January 2026. The page shows an empty cart with login prompts and notes that customers can cancel anytime before shipment, plus a brief explanation of what a pre-order is. Top 1 Comment Summary The author critiques a QVC product named after a US tech city, written by someone living near QVC’s home in a non-tech area of the US....