AI - Artificial intelligence

Unlike many startups producing nothing but vaporware, Jim Keller’s TensTorrent has actually delivered impressive-looking RISC-V AI accelerators, but there could be some trouble brewing. Starting with firmware version 19.5.0, the firm has now chopped the tensor core count on Blackhole p150 cards from 140 to 120, affecting both new cards and existing units already in customers’ hands.

Comments

Currently uses gpt5.2 for script generation, but the architecture is modular – you can swap in any local LLM (Qwen, Llama, etc.) if you want fully local.

Comments

The paper argues that we have been wasting a lot of expensive GPU cycles by forcing transformers to relearn static things like names or common phrases through deep computation. Standard models do not have a way to just look something up so they end up simulating memory by passing tokens through layer after layer of feed forward networks. DeepSeek introduced a module called Engram which adds a dedicated lookup step for local N-gram patterns. It acts like a new way to scale a model that is separate from the usual compute heavy Mixture of Experts approach.

Comments

Crossposted from https://fedia.io/m/pcmasterrace@lemmit.online/t/3295923

This paper is one of the more interesting takes on context extension I have seen in a while because it challenges the assumption that we need explicit positional encodings during inference. The authors make a case that embeddings like RoPE act more like scaffolding during construction rather than a permanent load bearing wall. The idea is that these embeddings are crucial for getting the model to converge and learn language structure initially, but they eventually turn into a hard constraint that prevents the model from generalizing to sequence lengths it has never seen before.

As stated in Goodhart’s law ‒ and excellently illustrated by autistic researchers’ darling xkcd‒ whenever a metric becomes a target, it stops being a good metric. Indeed, the system has grown so full of exploits that the target has been divorced entirely from quality.

instruct model https://huggingface.co/tencent/WeDLM-8B-Instruct

Sometimes AI need to be forced into actions, they block behaviour incorrectly identified as bad, unethical or evil. The methods to prevent such miss-classification by end-users will become harder with time.

Local LLMs are incredibly powerful tools, but it can be hard to put smaller models to good use in certain contexts. With fewer parameters, they often know less, though you can improve their capabilities with a search engine that’s accessible over MCP. As it turns out, though, you can host a 120B parameter model on a GPU with just 24GB of VRAM, paired with 64GB of regular system RAM, and it’s fast enough to be usable for voice assistants, smart home automation, and more. For reference, on 24GB of VRAM, the most practical dense model you’ll typically be able to fit will be a quantized 27 billion parameter model, accounting for the memory needed to hold the context window, too.

Archived

Blog post from the guy who created Dink Smallwood on why you shouldn’t feel bad for the Synths in “Detroit: Become Human” or feel bad about Cortana’s demise in Halo.

In software development, generating code hasn’t been the main bottleneck since we moved away from punch cards. The far bigger constraints are understanding the problem, communicating with stakeholders, working effectively with other people, designing the system, managing risks and trade-offs, and operating systems in complex social environments over time.

Start with what a reverse centaur is. In automation theory, a “centaur” is a person who is assisted by a machine. You’re a human head being carried around on a tireless robot body. Driving a car makes you a centaur, and so does using autocomplete.

A new Atomic macOS Stealer (AMOS) attack vector weaponizes Google searches and a user’s trust in AI chatbots, researchers have found. Once infected, the AMOS can collect data, passwords, and more from the infected Mac with alarming ease.

Two new open weight (MIT licensed) models from DeepSeek today: DeepSeek-V3.2 and DeepSeek-V3.2-Speciale, both 690GB, 685B parameters. Here’s the PDF tech report.

You can use Thaura for everyday tasks like writing emails, doing homework, and researching online. It remembers your conversations, helps you create documents and code, and even searches the web for you. And it works seamlessly with your existing tools through full OpenAI SDK compatibility.

Archived

In just one year, AI agents have gone from exploiting 2% of vulnerabilities in the post-March 2025 portion of our benchmark to 55.88%—a leap from $5,000 to $4.6 million in total exploit revenue. More than half of the blockchain exploits carried out in 2025—presumably by skilled human attackers—could have been executed autonomously by current AI agents.

DeepSeek has released V3.2, replacing the experimental version. There are two main models are open as always and can be downloaded from Hugging Face:

Note: this post is also applicable to AGENTS.md, the open-source equivalent of CLAUDE.md for agents and harnesses like OpenCode, Zed, Cursor and Codex.

President Donald Trump has issued a new Executive Order that launches the “Genesis Mission,” an AI-focused initiative that aims to make the “most complex and powerful scientific instrument ever built.”

Intel software engineers continue to be hard at work on LLM-Scaler as their solution for running vLLM on Intel GPUs in a Docker containerized environment. A new beta release of LLM-Scaler built around vLLM was released overnight with support for running more large language models.

Anthropic released Claude Opus 4.5 this morning, which they call “best model in the world for coding, agents, and computer use”. This is their attempt to retake the crown for best coding model after significant challenges from OpenAI’s GPT-5.1-Codex-Max and Google’s Gemini 3, both released within the past week!

Olmo is the LLM series from Ai2—the Allen institute for AI. Unlike most open weight models these are notable for including the full training data, training process and checkpoints along with those releases.