Sunday, 4 October 2026

40 stories · 1 from multiple sources

Run Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s

I wanted to see how far I could push a fairly ordinary laptop with a huge MoE model. Turns out, Qwen3.8 Flash Next 176B can run on: RTX 3080 Laptop — 16GB VRAM 32GB system RAM SSD No 128GB/256GB RAM workstation and no multi-GPU setup. I’m running it with TensorSharp , my open-source local LLM inference engine: TensorSharp on GitHub The interesting part for me wasn't simply getting a 176B model to…

Sunday, 4 October 2026

Run Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s

I asked Claude build a physically accurate O'Neill cylinder you can walk around

Remove and Disable Apple Macos27 AI Models Tool

More stories

Homa: The End of TCP for AI Clusters [video]

Building a RAG Pipeline for Semantic Code Search

What I learnt co-leading an AI Safety bootcamp for legal and governance practit

Micron CEO Says Memory Supply Will Be Much Tighter in 2027 and 2028 Than in 2026

LLM Inference Dashboard

Update #4: Post training yandex/AliceAI-80B-A3B [instruct!] from scratch

Also today

Benchmarking decision models is fun - Clef Q8 vs Jev

Built a quick, sub-15ms Rust CLI/TUI to pack repos into prompts without burning 40k tokens on lockfiles and junk

Qwen3.5 arch implementation in FPGA fabric for 9B/27B INT4 models on relatively cheap eBay mining hardware

How to scale intent, quality, and artistry with AI [video]

I trained a 3.87B MoE (1.45B active) from scratch on only 86.5B tokens

How long before we have a local model capable of modeling?

Google froze its open source bug bounty program due to a ‘significant rise’ in AI submissions

Can ‘super intelligence’ and a non-binding safety pact solve AI’s image problem?

Sam Altman's sister amends lawsuit accusing OpenAI CEO of sexual abuse

I Quit OpenAI Because Its Culture Is Broken

RuneScape's Position on Gen AI

🥇Top AI Papers of the Week

From 1x3090 to 20 DGX Sparks: my house fuses were the first bottleneck

We're going to need default hard budget caps on pretty much everything

The Agent Said It Was Done. The Database Disagreed.

Well, if AI said it, it must be true

An AI couldn’t beat humans at StarCraft, so it decided to cheat

Trump unveils his new Super Intelligence Force

Rex's Dino Store

Sellio — AI customer support in one shared inbox powered by AI agents

Aperture — An AI code editor that checks its own work

🤖 AI Agents Weekly: Gemini 4 Argon, OpenAI Dots, GPT-6.1 Sol, Claude Sonnet 5.5, Pi 1.0, Claude Code Mods, and More

[AINews] not much happened today

Amazon responds to data center backlash, says it no longer uses NDAs

Thinking Orbs — Animated AI Status Indicator Component Library for React

Capcom is preparing for a ‘future where we create games together with AI’

OpenAI safety employee resigns, claiming the company’s ‘culture is broken’

Splice CEO Kakul Srivastava thinks AI emails are killing conversations

An OpenAI safety employee has quit and is sounding the alarm

All the AI agents that can live in your text messages

Sapien — AI market research with synthetic populations