Tencent says its Hy4 joins 'top-tier' of open-source AI models
Nikkei Asia — Chinese tech giant says latest model is competitive with Anthropic, OpenAI releases
Models & Research
Artificial intelligence models and research encompass the foundational algorithms, architectures, and scientific studies that drive system capabilities. Truncated covers new model releases, underlying technical advances, and published research papers.
Qwen 3.8 27B on my laptop can interpret model weights, too
Simon Willison’s Weblog — Yesterday’s hot paper was Sparks of Weight-Space Programming: LLMs can natively read & write neural network weights by Rikiya Takehi, Jayden Teoh, Alex Zhang, and Omar Khattab. The paper found …
Election jitters, self-improving AI, and Wall Street tantrums · Cautious Optimism
Cautious Optimism — What, me worry?
John Ternus talks plans to improve Siri AI
9to5Mac — Apple CEO John Ternus was asked about Siri AI’s future, here’s what he said about the company’s plans to improve and expand its assistant.
Gemini 4 Carbon: our next frontier model
Google — Gemini 4 Carbon is our strongest model with state-of-the-art cybersecurity and long horizon capabilities.
Meituan Releases LongCat 2.5 on Chinese Chips; Benchmark Claims Await Independent Tests
Tech Times — Meituan LongCat 2.5 launches November 2, 2026 as a 1.6-trillion-parameter multimodal AI model designed for autonomous coding agents, built for a second consecutive training run entirely on domestic Chinese chips. Meituan claims performance between Claude Opus 4.8 and Opus 5, but no independent
ChatGPT Dots are getting smarter, and you won't have to lift a finger
Android Authority — OpenAI is making ChatGPT Dots smarter with GPT-6 Astra and GPT-6.1 Terra, allowing them to automatically pick the best AI model for each task.
Nace AI has launched Drex for Finance, a model designed for split-second financial tasks, like determining whether a ca…
Claude Fable 5.5, GPT 6.2 Astra Could Be Right Around the Corner, Leaks Suggest
Decrypt — Users say Claude is routing some requests to a stronger unreleased model, and a Mythos 5.5 label appeared in Anthropic’s site code.
LLMs can read & write neural network weights: "Weight Space Programming" cuts Autoresearch out of the RSI loop
Latent Space — Editing Qwen’s weights to reduce reasoning loops, DeepSeek’s super-agent prototype, and the GDPx benchmark.
Progress on provable inference for large language models
anthropic.com — Sharing progress on a powerful type of cryptographic proof known as a zkSNARK, which is uniquely fit for provable inference.
Haiku 5.5 now available on Kiro in AWS GovCloud (US) - AWS
How We're Training a Figma-Fluent Editing Model
Figma — Here, we’re sharing the research behind how we’re translating that to design editing workflows.
A Forgotten 1949 Memo, A 1957 Map Of Meaning, And The New Proof That Separately Trained AIs Can Learn The Same World.
brianroemmele.substack.com — A MEMO FROM CARLSBAD
Building a multiplayer game world with autonomous Fable-class models
simonwillison.net — Don't be put off by the clickbait headline, this piece by Jacob Sansbury describing how they are building Spawn is a fascinating exploration of the bleeding edge of what's possible …
The most important part of AI chatbots is being ignored
New Scientist — AI firms can increase the power of their chatbots without updating to new models. Columnist Jacob Aron explains how a simple bit of software can give AI a boost – and whether AI will soon be able to rewrite that code itself
Our cyber models are getting really strong.
NVIDIA Nemotron 3.5 Transcribe and Granary v2: Intelligent Multilingual Transcription You Can Prompt and Adapt
huggingface.co — A Blog post by NVIDIA on Hugging Face
Build specialized multimodal models with LFM3
huggingface.co — A Blog post by Liquid AI on Hugging Face
Decisions for a penny: introducing joey-1, our fastest decision model
Baseten — A week ago we shipped mammoth-1 (27B) and elephant-1 (9B), the #1-ranked decision models. Today we’re adding joey-1, our fastest, cheapest 2B decision model.
Dismiss email signup
Subscribe to 
The top AI news in your email delivered each morning. Choose categories