Qwen 3.8 27B

505 erdaltoprak 2026-08-14 15:21:20 UTC huggingface.co Llama Share 348

This repository provides FP8-quantized model weights and configuration files for the post-trained Qwen3.8-27B model, offering managed inference without infrastructure maintenance through the Qwen Cloud service. The model supports various tasks, including vision-language understanding, agent execution, and long-horizon agentic tasks.

Count Binface receives over a quarter of votes in Clacton by-election

277 tcp_handshaker 2026-08-14 17:21:20 UTC bbc.com Llama Share 178

Count Binface, a comedian and satirist, received nearly 10,000 votes in the Clacton by-election. He has previously run for election as Lord Buckethead and won viral fame with his novelty candidacy.

Seven books I keep close because I love them

193 surprisetalk 2026-08-14 15:21:20 UTC blog.plover.com Llama Share 84

The author has a collection of books on their shelf, including Roget's Thesaurus, Sir Thomas Browne's works and the Decameron. They reflect on how these books inspire and inform their writing.

RustDesk now supports true unattended remote access on Wayland

111 rustdesk 2026-08-14 16:21:20 UTC rustdesk.com Llama Share 48

RustDesk now offers unattended access on Wayland without approval, supporting multi-monitor setups. A preview build is available for x86_64 Debian/Ubuntu-based systems to test and provide feedback.

Why does Opus 5 feel worse to work with?

547 numeri 2026-08-14 10:21:20 UTC mun-logadan.github.io Llama Share 506

The author believes Opus 5 is a capable model but feels like a downgrade due to its tendency to make bold assumptions in the face of ambiguity, prioritizing benchmark performance over nuanced understanding. This approach may not be suitable for real-life applications where agents need to ask questions and seek clarification.

GLM-5.3: Frontier coding with emergent cyber capabilities

958 pella 2026-08-14 06:21:20 UTC z.ai Full Share 477

Google is making private AI practical with homomorphic encryption

111 u1hcw9nx 2026-08-14 16:21:20 UTC blog.google Llama Share 72

Google's HEIR compiler toolchain enables secure private AI inference using homomorphic encryption, balancing privacy and security. The open-source tool converts pre-trained models to operate on encrypted inputs, making it a one-click solution for non-experts.

Introducing Toast 1

126 mplappert 2026-08-14 15:21:20 UTC mixedbread.com Llama Share 38

Toast 1, a specialized search agent, offers frontier search quality at up to 10x cheaper and 12x faster than existing models. It can be used as a standalone model or as a subagent with other models.

Maximizing the value of your Claude Code sessions

69 twapi 2026-08-14 16:21:20 UTC claude.com Llama Share 37

Claude Code provides tips to optimize token usage and session efficiency, including setting models and effort levels before starting, using /context to view loaded context, compacting conversations when taking breaks.

AI by Hand

79 sans_souse 2026-08-14 16:21:20 UTC byhand.ai Full Share 9

Math, Algorithms, Architectures, by hand. Click to read AI by Hand ✍️, by Prof. Tom Yeh, a Substack publication with tens of thousands of subscribers.

Show HN: Embed a real Linux terminal on your website

9 raboukhalil 2026-08-14 18:53:20 UTC sandbox.bio LocalLLM Share 5

Sandbox.bio allows embedding a terminal on websites via script tags or Markdown. Config files and preloaded files can be specified for customization.

Ultraviolet Bird Photography

55 EndXA 2026-08-14 16:21:20 UTC uvbirds.com Llama Share 8

Birds see colors that humans can't, including ultraviolet light, which affects their plumage and behavior. This hidden dimension of color helps birds recognize species and navigate their environment.

I turned my RSS feeds into an e-ink newspaper to stop reading on my phone

81 speckx 2026-08-14 15:21:20 UTC heyjonny.dev Llama Share 33

The author discovered a pocket-sized e-reader called the X4, which they customized with open source firmware Crosspoint to improve reading experience. They created their own "newspaper" using an app called Feedpaper, allowing them to read personal blogs on the device.

Every Fucking Website (2020)

609 doubletwoyou 2026-08-14 15:21:20 UTC lxe.github.io Llama Share 351

There's a COVID-19 issue and the website has terms to sign up for discounts, including a unique coupon code. Users must agree to these terms before getting 10% off.

Moving integer division to floating-point is trivial

14 matt_d 2026-08-14 18:21:20 UTC marc-b-reynolds.github.io Llama Share 6

The author claims that for two integers x & y (with 53/24 bits for double/single precision), promoted to floating point, integer division can be performed using floating-point operations with no rounding errors. The key observation is that there are no ties in the midpoint between two floating-point numbers, so only round-to-nearest applies and never causes rounding errors.

DeepSeek peak/off-peak pricing update

219 fagnerbrack 2026-08-14 10:21:20 UTC api-docs.deepseek.com Llama Share 117

V4-Pro and V4-Flash models have varying reasoning efforts for tasks, with Expert Mode available on app/web. API pricing is updated with peak and off-peak rates, offering more flexible workload scheduling options.

France's top court blocks social media ban for under-15s

141 BlueBerry2001 2026-08-14 16:21:20 UTC reuters.com Full Share 80

Please enable JS and disable any ad blocker

Don't classify, hallucinate

181 softwaredoug 2026-08-14 11:21:20 UTC softwaredoug.com Llama Share 74

This text discusses using LLMs to classify products, specifically with the Wayfair WANDS e-commerce dataset. It proposes a method of asking an LLM to generate fake classifications and then finding the most similar real ones through dot product calculation.

Forbidden City Cabbage

31 dude250711 2026-08-14 16:21:20 UTC cnn.com Llama Share 4

The National Palace Museum in Taiwan features a jadeite cabbage sculpture, considered one of its most popular and iconic items. The cabbage's enduring fame may be due to its perceived symbolism and craftsmanship, as well as its connection to the imperial court.

Show HN: LuaCAD – Parametric CAD Scripted in Lua

37 adius 2026-08-14 17:21:20 UTC luacad.ad-si.com Llama Share 6

LuaCAD is a scriptable CAD tool with Lua, allowing users to create parametric 2D and 3D models that can be exported in various formats. It supports CSG operations and rendering, with options for exporting to different file formats or using an installed OpenSCAD binary.

A Contract-Grade Verifier for LLM-Generated GPU Kernels

12 Jimmc414 2026-08-14 17:21:20 UTC arxiv.org Llama Share 0

Researchers verified 2,638 machine-generated GPU kernels using a contract-grade verifier and found 39.5% to be broken beyond any tolerance argument. The findings suggest that the field's standard test may overestimate kernel correctness rates by up to 62%.

RayforceDB – a pure C analytics database with a Lisp-like syntax

26 sourdecor 2026-08-14 16:21:20 UTC rayforcedb.com Llama Share 14

Rayforce combines columnar analytics, graph traversal, and recursive queries into a single embeddable execution pipeline for teams measuring latency in microseconds. It offers a unified analytics and graph pipeline with a managed cloud experience for seamless production workloads.

In Australia, a home battery boom has helped cut wholesale power prices

254 speckx 2026-08-14 14:21:20 UTC e360.yale.edu Llama Share 194

Australia's home battery subsidy program has led to over 500,000 installations, slashing wholesale power prices by roughly half. The program is helping households use more solar energy and reducing grid costs nationwide.

Show HN: Mole – Deep research agent for your terminal

6 lajosdeme 2026-08-14 18:54:20 UTC github.com/lajosdeme LocalLLM Share 0

A deep-research agent called mole enforces budgets, verifies quotes, and keeps data local. It uses MCP for interaction, runs as a static binary, and ensures claims are backed by verbatim sources.

AI Model Atlas – visualizing populations of ML models as interconnected 3D graph

36 bj-rn 2026-08-14 15:21:20 UTC run.cosmograph.app Full Share 5

Gödel, Escher, Elisp: The Beauty of Macros

62 JNRowe 2026-08-14 14:21:20 UTC chiply.dev Llama Share 19

Emacs Lisp macros allow users to leverage the blur between program and data, enabling flexible and expressive code. Macros can be used to extend the language, adding new defining forms that communicate in the problem domain's own diction.

Self-hosted web push Cloudflare Worker, works on iOS

43 saiday 2026-08-14 15:21:20 UTC kukuroo.cc Llama Share 15

Notifications should be sent directly to a user's device via Cloudflare Workers, bypassing middlemen like apps and servers. This approach allows for secure, encrypted alerts without requiring an app or dashboard access.

The TEMU-Fication of Software, Digital Goods and Services

98 surprisetalk 2026-08-14 12:21:20 UTC xn--gckvb8fzb.com Llama Share 74

The author speculates that the future of software, books, music, and movies will have a two-tier market system where most content is cheaply generated by AI but high-quality human-made work becomes a luxury segment. The bulk of the market will be driven by ultra-cheap AI-generated content while a smaller niche will value and pay for human-made work.

Earth.nullschool.net

37 TremendousJudge 2026-08-14 15:21:20 UTC earth.nullschool.net Llama Share 14

The National Oceanic and Atmospheric Administration (NOAA) is facing mass firings and budget cuts, which could disrupt weather forecasting services. Users are advised to contact representatives or share concerns on social media to advocate for the reversal of these actions.

HashAgent – Share an AI agent as a URL, runs locally via WebGPU

38 masonhsu 2026-08-14 13:21:20 UTC hashagent.pages.dev Full Share 4

Create and share private AI agents as URLs. Local WebGPU inference, built-in web search, no account or tracking.