Qwen 3.8 27B

280 erdaltoprak 2026-08-14 15:41:57 UTC huggingface.co Llama Share 209

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.

Seven books I keep close because I love them

136 surprisetalk 2026-08-14 15:41:57 UTC blog.plover.com Llama Share 46

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.

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

45 tcp_handshaker 2026-08-14 16:46:57 UTC bbc.com Llama Share 9

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.

RustDesk now supports true unattended remote access on Wayland

54 rustdesk 2026-08-14 16:41:57 UTC rustdesk.com Llama Share 14

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.

Introducing Toast 1

89 mplappert 2026-08-14 15:41:57 UTC mixedbread.com Llama Share 25

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.

Why does Opus 5 feel worse to work with?

426 numeri 2026-08-14 10:41:57 UTC mun-logadan.github.io Llama Share 410

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

919 pella 2026-08-14 05:41:57 UTC z.ai Full Share 465

AI by Hand

32 sans_souse 2026-08-14 16:41:57 UTC byhand.ai Full Share 4

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

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

19 sourdecor 2026-08-14 16:41:57 UTC rayforcedb.com Llama Share 11

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.

Ultraviolet Bird Photography

27 EndXA 2026-08-14 15:41:57 UTC uvbirds.com Llama Share 2

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.

Every Fucking Website (2020)

541 doubletwoyou 2026-08-14 14:41:57 UTC lxe.github.io Llama Share 316

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.

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

57 speckx 2026-08-14 14:41:57 UTC heyjonny.dev Llama Share 24

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.

Forbidden City Cabbage

13 dude250711 2026-08-14 16:41:57 UTC cnn.com Llama Share 0

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.

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

100 BlueBerry2001 2026-08-14 16:41:57 UTC reuters.com Full Share 35

Please enable JS and disable any ad blocker

Maximizing the value of your Claude Code sessions

18 twapi 2026-08-14 16:41:57 UTC claude.com Llama Share 2

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.

Discrete Fourier Transform by Hand

9 Bluestein 2026-08-14 16:53:57 UTC byhand.ai LocalLLM Share 1

DFT simplifies to matrix multiplication, revealing its structure as a series of cosine wave projections. Neural networks like U-Net learn similar transformations, while DFT remains a fixed, manually designed method for signal analysis.

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

218 speckx 2026-08-14 14:41:57 UTC e360.yale.edu Llama Share 165

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.

Google is making private AI practical with homomorphic encryption

67 u1hcw9nx 2026-08-14 16:41:57 UTC blog.google Llama Share 49

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.

Don't classify, hallucinate

167 softwaredoug 2026-08-14 11:41:57 UTC softwaredoug.com Llama Share 71

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.

DeepSeek peak/off-peak pricing update

203 fagnerbrack 2026-08-14 10:41:57 UTC api-docs.deepseek.com Llama Share 108

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.

"Solving a largely imaginary user goal"

35 euthymiclabs 2026-08-14 16:41:57 UTC unsung.aresluna.org Llama Share 37

The author agrees with Lea Verou's proposal for a "smart" dual-state dark mode toggle, which simplifies the user interface by hiding complexity. This approach abstracts away underlying system complexities and focuses on user goals, reducing cognitive load and improving overall usability.

Self-hosted web push Cloudflare Worker, works on iOS

35 saiday 2026-08-14 15:41:57 UTC kukuroo.cc Llama Share 11

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.

A Contract-Grade Verifier for LLM-Generated GPU Kernels

4 Jimmc414 2026-08-14 16:58:57 UTC arxiv.org Prefix Share 0

Systems that generate GPU kernels with language models report high correctness rates. Those rates come from a single loose test: run the kernel on a few random inputs at one fixed shape and accept it if the output is close to a reference. A kernel can pass that test and still be silently wrong. It can return an ordinary number where the true answer is a NaN or an infinity, differ from run to ...

Cursor is now a part of SpaceX

60 billy_jones_75 2026-08-14 13:41:57 UTC cursor.com Llama Share 40

Cursor has been acquired by SpaceX, gaining access to the largest fleet of GPUs. This acquisition will enable Cursor to build stronger models at lower cost, making it easier for users to solve harder problems.

The TEMU-Fication of Software, Digital Goods and Services

77 surprisetalk 2026-08-14 12:41:57 UTC xn--gckvb8fzb.com Llama Share 57

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.

Show HN: LuaCAD – Parametric CAD Scripted in Lua

6 adius 2026-08-14 16:44:57 UTC luacad.ad-si.com LocalLLM Share 0

LuaCAD allows parametric 3D modeling in Lua, exporting to multiple formats or rendering to PNG. It uses Manifold for CSG operations or generates OpenSCAD code, with full Belfry library support.

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

33 masonhsu 2026-08-14 13:41:57 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.

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

23 bj-rn 2026-08-14 14:41:57 UTC run.cosmograph.app Full Share 4

Gödel, Escher, Elisp: The Beauty of Macros

34 JNRowe 2026-08-14 14:41:57 UTC chiply.dev Llama Share 11

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.

Earth.nullschool.net

22 TremendousJudge 2026-08-14 15:41:57 UTC earth.nullschool.net Llama Share 12

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.