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
IBM Research Zurich’s open-source Docling project is building a practical bridge between unruly documents and AI systems. It parses PDFs, Office files, audio and images into a unified representation, with layout analysis, OCR and visual-language-model support built in. The useful part is not that another tool has discovered…
Read the report & discussion ↗Watch Desk
Jared Palmer has released Kev, a family of small AI decision models that return answers such as yes/no probabilities, multiple-choice decisions and ratings instead of conversational prose. The practical point is compellingly unglamorous: developers can run these models locally for fast, repeated decisions without sending every request to…
Microsoft AI Watch
Microsoft says its open-source HASTE system is helping emergency responders identify damaged areas, blocked routes and likely aid priorities within hours rather than the days or weeks a manual assessment can take. The practical point is simple: faster maps can help people decide where to send scarce rescue…
Watch Desk
Snorkel AI has raised $350m at a $3.5bn valuation, according to Reuters, putting a substantial price tag on the less glamorous work that makes AI systems useful: creating and checking the data they learn from. The company says its platform uses people and AI agents to develop and…
Watch Desk
Apple is positioning its new Mac mini and Mac Studio computers as local AI machines for corporate buyers, taking aim at the cloud costs and hardware dominance of Microsoft and NVIDIA. The important shift is practical: companies could run demanding AI work on equipment they own, rather than…
Watch Desk
Alphabet’s robotics company Intrinsic has open-sourced key parts of its software platform, including tools for real-time control, pose estimation, motion planning and camera calibration. The release matters because it gives developers reusable building blocks for industrial robots, rather than another isolated demonstration with a suspiciously cinematic soundtrack.
Watch Desk
OpenRouter has launched a Batch API that cuts inference prices by half for users willing to let requests run asynchronously within a 24-hour window. The service is aimed at large, non-urgent workloads such as bulk text generation and embeddings, where immediate responses are less valuable than a lower…
Watch Desk
AI sovereignty is not achieved by swapping one foreign chatbot for another, argues an O’Reilly Radar essay published on 22 September. Its central point is sharper: governments and institutions need open infrastructure across the AI stack if they want genuine control and the ability to change providers.
Watch Desk
Machine-learning systems can quietly become less reliable even when their code, weights and confidence scores stay exactly the same. A new explainer by Cristobal Santana sets out how changing data distributions can cause that silent failure, and why checking the data matters as much as checking the model.
Watch Desk
Yandex has released the weights for Alice AI Foundation, an 80bn-parameter language model that the company says is available for commercial and research use. It is a substantial open-model launch, but also a revealing bid to build a more self-sufficient Russian AI stack.
Alibaba Qwen Watch
An exploratory study suggests Qwen models may contain internal representations of older language models, including GPT-2. The intriguing finding is not that Qwen can imitate GPT-2, but that its completions may track GPT-2’s hidden continuation more closely than Qwen’s own usual output.
Watch Desk
StepFun has released Step 5 Preview, a large mixture-of-experts model offering a one-million-token context window, image input and API pricing of $1 per million input tokens and $2.70 per million output tokens. The practical significance is not merely that another frontier model has arrived. It is that long-context,…
Watch Desk
A new tool called Product Prover applies formal-verification ideas to ordinary product specifications, using an LLM to reconstruct state machines and flag missing rules before developers generate code. That makes it a useful AI engineering story because it targets one of software’s oldest problems: building precisely what nobody…
Watch Desk
South Korean semiconductor-equipment maker Melcon plans a KOSDAQ listing on 15 October, aiming to raise up to 30.8 billion won for High-NA EUV development, extra production capacity and expansion into TSMC and HBM-related work. It is a useful reminder that the AI hardware race also depends on the…
Watch Desk
Researchers have introduced RRSI, a method for automatically improving the software harness around AI agents while trying to stop those agents overfitting to the tasks used for training. The important shift is that the work targets the tools, prompts and control logic surrounding a model, not just the…
Watch Desk
A new benchmark finds that even advanced computer-use agents can reach the wrong result by taking the wrong route. OSWorld-Pro evaluates more than 300 tasks and 2,800 subgoals, exposing the small, consequential mistakes hidden by a tidy final answer.
Hugging Face Watch
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…
