From the archive
AI models are the systems behind chatbots, coding assistants and tools that work with text, images, audio and video. Follow new language model releases, open-weight models, local AI and benchmark results. Our coverage looks at capabilities, context limits, access, pricing and evaluation methods so you can compare the evidence behind a launch, not just the claims in its announcement.
Lead story
Hugging Face has hired Jun Kim, the creator and maintainer of oMLX, to support the MLX community and help move local AI development on Apple Silicon from promising side project to properly funded work. The important detail is continuity: Kim will keep leading oMLX, while Hugging Face says…
Read the report & discussion ↗Watch Desk
AI agents that rewrite their own instruction and memory files can quietly degrade their performance, according to Jamie Watters. The practical fix is decidedly unglamorous: keep a person in the editing loop, review the files regularly and delete the bits that no longer earn their keep.
DeepSeek Watch
DeepSeek researchers have introduced DSec, a production sandbox platform designed to run large-scale training and evaluation for AI agents. Its significance is less glamorous than a new model name, but arguably more useful: DSec targets the infrastructure bottlenecks that appear when agents need vast numbers of isolated environments…
Watch Desk
Compute:Arena has launched a public leaderboard and command-line tool for comparing local AI inference across models, quantisations, chips and runtimes. Its useful promise is simple: replace spec-sheet optimism with measurements tied to the exact model file and hardware combination.
Watch Desk
AI code reviewers found more real bugs when they used open-source review harnesses rather than relying on a single prompt, according to a new research evaluation. The catch is equally useful: better coverage took more time, tokens and money, while open-weight models delivered the strongest value for the…
OpenAI Watch
Researchers report that GPT-6 Astra can control robots directly on a range of tasks, outperforming the other public models tested. The more useful finding is less cinematic: Astra appears capable of understanding what a robot should do, but remains unreliable when the job demands precise, dynamic or two-handed…
Watch Desk
Researchers have introduced CausalWM, a 16-billion-parameter embodied world model that reasons about cause and effect before predicting what happens next. The work matters because it aims to move AI systems beyond spotting visual patterns and towards modelling how actions change the physical world, a rather important distinction once…
Watch Desk
Fireworks AI is arguing that the next performance leap in coding agents may come from routing each task to the right model, rather than asking one frontier model to do everything. Its new FireRouter approach keeps a pool of open and closed models available, while preserving context caches…
Watch Desk
A study of an AI coding agent found that describing the desired structure in plain language, without naming the scoring formula, improved results and reduced correctness losses. The trade-off was a roughly 20% increase in both agent costs and tool turns, a neat reminder that clever prompting can…
Watch Desk
Google has introduced GKE Pod snapshots, a new Kubernetes feature that saves a running workload’s state, including CPU and GPU memory, so it can be restored on demand. The practical prize is faster AI inference startup, with Google claiming latency reductions of up to 89% and a route…
Watch Desk
TinyTorch is a new free, open-source curriculum from PyTorch that takes learners from basic tensors to transformer models by having them build a working machine-learning framework in pure Python. It runs on an ordinary laptop without a GPU, making the machinery behind modern AI considerably less mysterious and…
Open WebUI Watch
Open WebUI 0.11.4 adds a substantial set of terminal and agent features, including automatic skill discovery, reusable skills created from chats, folder uploads, terminal tabs and side-by-side file comparison. It also ships a smaller container image and security fixes, making this more than a cosmetic release for people…
AWS AI Watch
xAI’s Grok 4.6 is now available through Amazon Bedrock, giving organisations a new route to a model aimed at long-running agents, coding and knowledge work. The practical change is not just another model in a catalogue: Bedrock users can choose between endpoints, reasoning levels, cross-Region routing, prompt caching…
Watch Desk
A Hugging Face Community Article describes Jevini, a system for turning repeated business choices into versioned decision circuits rather than asking a large language model to improvise every time. The practical idea is simple but useful: let a larger reasoning model design the process, then use a smaller…
Watch Desk
Tony Blair’s institute is calling for governments to build a serious independent testing and certification industry for AI, rather than responding to public anxiety by either freezing adoption or waving it through. The proposal matters because it puts accountability, liability and practical checks at the centre of the…
OpenAI Watch
OpenAI has launched Astra for Law, a legal-focused version of its GPT-6 Astra model, initially for selected law firms and with API access planned later. The important shift is that OpenAI is not merely offering lawyers a clever chatbot: it is pairing a frontier model with a large…
Watch Desk
ASUS has unveiled the ProArt Air, a 1.28kg business laptop designed to run AI-agent tools with local Intel processing and Tencent’s WorkBuddy platform. The practical shift is modest but real: AI features are being bundled into the office computer rather than left as a cloud tab someone has…
