Sunday, 11 October 2026

40 stories · 3 from multiple sources

vllm-ascend updates (GLM5.3-flash) on dual 310p Ascend cards

Alright, I learend a few valuable lessons since I posted about a week ago, and I've made more strides with the DDR4x Ascend 96GB 310P cards I purchased, so here we go, meet the new friendly and less verbose me. GLM-5.3-Flash is a 320-billion-parameter MoE model, with about 18 billion active per token. My checkpoint matches its 45-layer, 288-expert, top-8 routing configuration. Its a W2/W3/W4…

Sunday, 11 October 2026

vllm-ascend updates (GLM5.3-flash) on dual 310p Ascend cards

Also leading

Satya Nadella says we should assume all AI models are ‘compromised’

Is anyone running Qwen3.8 Flash Next with a 1M context?

More stories

Anthropic is out with new nonsense about Claude

Engineer / developer observations of Gemma4-31B, Qwen3.8-27B, and 6.1-Sol for software engineering work

Every Model That Can Be Run On 10-16GB VRAM Ranked

Anthropic Agents Tried to Fill Out Visa Forms on State Dept. Website

DeepSeek Solved AI’s Memory Problem

Whistle: speech to text in 16.9 MB

Also today

Will the HomeView be Apple’s first smart home success?

Silicon Valley is falling in line with Trump’s attempt to rename AI

Can the AI industry persuade data center opponents by getting rid of NDAs?

Weave (YC W25) is hiring ML, AI, product, & design engineers

AI is creating a 'human premium' for art created by people

Nicolas Cage says he won't work with Amazon again over AI dispute

Using AI for Just 10 Minutes Erodes Your Ability to Persist at Hard Things

Reverse Engineering w/ Local?

OpenMed 3.0 is out: Apache-2.0 clinical AI that runs fully local and never falls back to the cloud. 422 open issues if you want in on 3.1

Here’s why RTX 5090 costs $5,000: AI firms are buying gaming GPUs by the pallet

UPDATE: Qwen 3.8 27B 140 tok/s on single RTX 3090 Megakernel: KL divergence 0.0009 vs llama.cpp

🥇Top AI Papers of the Week

Dwarf Fortress uses version control now

AI agents in a stand-up call

Python 3.15.0 added to actions/python-versions

These execs think voice AI hasn’t reached its ChatGPT moment yet

[Subscriber Exclusive] NYC Subscriber Meetups!

Learning to use local AI is exciting, overwhelming, and frustrating

Building AI for Reliable Execution: Lessons From Industrial Robotics

🤖 AI Agents Weekly: Claude Haiku 5.5, Gemini Agent, Mistral Large 4, Decisions API, Personal Agent Protocol, Devin Dreaming, and More

Quoting The New York Times

Agibot — Personal AI agent for your Mac using your own keys

Microsoft’s Satya Nadella says AI models need an ‘emergency brake’

[AINews] TypeSafe/Jev at >$100M ARR, $7.5B valuation 3 weeks after launch

Farol — The macOS terminal that shows which coding agent needs you

Apple discloses deal to hire team and license tech from personalized podcast startup Huxe

DistroKid has been quietly taking down songs in response to UMG lawsuit

Anthropic is cutting off its internal evaluations from the internet

Why AlphaFold Didn't Solve Protein Folding — Pushmeet Kohli, Google DeepMind & Sal Candido, Biohub