🧠 Large Language Models

News about foundation models, LLMs, multimodal models, benchmarks, and model releases from OpenAI, Anthropic, Google, Meta, Mistral, and others.

1,682 articles

Mistral's paradox: Europe's tech sovereignty bet leans on China's Z.ai

Mistral, one of Europe's leading AI companies, is offering access to Z.ai's GLM-5.2 model, exposing a tension between European sovereignty ambitions and commercial reality. The move shows how business incentives can clash with political goals in the AI race. Source: scmp.com

Study finds LLMs increasingly used to boost autonomous driving systems

A new survey highlights how large language models are being applied across self-driving technology. One common approach breaks the driving task into separate modules covering perception, prediction, and planning, including route and behavior decisions. Source: eurekalert.org

Alibaba's Qwen3 combines deep reasoning with faster responses

Qwen3 is a series of large language models built by Alibaba Cloud's Qwen team, aimed at powering a wide range of AI applications. The lineup is designed to balance stronger reasoning ability with quicker response times. Source: trendhunter.com

Study tests LLM-based clinical decision support in the ER

A new prospective study evaluates how a large language model performs as a clinical decision support tool in emergency department settings. Researchers note that solid real-world evidence for AI-driven decision support in emergency medicine has so far been scarce. Source: nature.com

Tesla adds ByteDance's Doubao AI model to cars sold in China

Tesla has integrated Doubao, a large language model built by ByteDance, into the in-vehicle systems of its electric cars sold in the Chinese market. The move brings a locally developed LLM into Tesla's onboard AI features for Chinese customers. Source: china.org.cn

Why large language models must also serve small language communities

While much attention focuses on AI's security and military risks, experts warn that large language models pose a less-discussed cultural and economic threat to smaller language communities. Without dedicated LLM support, minority languages risk being marginalized in the AI-driven digital economy. Source: businesstimes.com.sg

Researchers probe whether LLMs show signs of consciousness

In a recent test on Anthropic's Claude Sonnet 4.5 model, researchers asked the LLM to count to five while simultaneously performing another task, aiming to explore how it handles self-referential prompts. The experiment is part of a broader effort to understand whether such behaviors hint at forms of self-awareness in large language models. Source: economist.com

FTC Chair Flags Antitrust Risks Brewing in AI Supply Chain

FTC Chairman Andrew Ferguson says competition among large language model developers is 'pretty ferocious,' but warns that antitrust risks could emerge further upstream in the AI supply chain. Source: broadbandbreakfast.com

1-Bit LLMs Aim to Bring 27B-Parameter AI Power to Phones and Laptops

While open-weight language models keep growing in size, a parallel trend explores extreme compression—using 1-bit models to squeeze capabilities comparable to 27-billion-parameter models onto everyday devices. Source: intelligentliving.co

South Korea Launches First Sovereign AI Appliance with Local Chip and LLM

KT's new NPU LLM Station pairs Rebellions' ATOM-MAX inference chip with the Mi:dm K 2.5 Pro language model in a single on-premises server, marking Korea's first fully domestic sovereign AI appliance. Source: techtimes.com

US Army AI Center Teams Up with EdgeRunner AI on New Local LLM

The Army's Pittsburgh-based AI center is partnering with Washington-based EdgeRunner AI to build a language model that can run locally, cutting reliance on the cloud and reducing cybersecurity exposure. Source: bizjournals.com

Study Examines Safety and Security Challenges of LLMs in Healthcare

A new analysis looks at how generative AI and large language models are rapidly being adopted in clinical care, while highlighting the safety and security concerns this integration raises. Source: nature.com

LLM-Generated Compilers Speed Up Trapped-Ion Quantum Computing

Researchers used a large language model to automatically generate compilers for shuttling operations in trapped-ion quantum computers, cutting the required timesteps by up to 76%. Source: quantumzeitgeist.com

Apple Reportedly Building China-Specific AI Model Ahead of Apple Intelligence Rollout

According to three sources familiar with the matter, Apple has trained a dedicated large language model tailored to the Chinese market, working with Alibaba for support. The move comes as Apple prepares to bring its Apple Intelligence features to Chinese users. Source: hungarianconservative.com

EdgeRunner Teams Up With U.S. Army AI Center to Develop Military-Specific LLM

EdgeRunner is partnering with the U.S. Army Artificial Intelligence Integration Center to build a large language model tailored for Army needs. The project focuses on open-weight models that can run locally, offering better performance, efficiency, and control for sensitive national security applications. Source: 01net.it

Baidu Chief Vows Ernie Model Comeback After Revenue Miss Hits Stock

Baidu CEO Robin Li pledged on Tuesday to restore the company's Ernie large language model to the cutting edge of AI development. The promise follows a disappointing revenue report that sent Baidu's shares tumbling. Source: kfgo.com

DARPA Official: Large Language Models Still Fall Short for Military Use

Speaking in Alexandria, Virginia, a DARPA official said that despite the hype around AI tools like large language models, they currently offer limited practical value for military missions on their own. The official emphasized that LLMs need to be paired with other capabilities to be truly useful in defense applications. Source: nationaldefensemagazine.org

New AI Model Explains Its Medical Image Diagnoses Using Concepts

Researchers built a biomedical foundation model that combines vision and language pretraining with explicit medical concepts, aiming to make AI diagnoses more interpretable for doctors. The approach addresses a key limitation of current multimodal medical AI systems, which often lack transparency. Source: nature.com

ByteDance and Tsinghua Unveil CUDA Agent for Automated GPU Kernel Writing

ByteDance Seed and Tsinghua AIR introduced CUDA Agent, a large-scale reinforcement learning system that trains an LLM to automatically generate CUDA GPU kernels. The project aims to speed up and automate low-level GPU code optimization typically done by human engineers. Source: marktechpost.com

MIT Study Examines How ChatGPT Affects Brain Activity While Writing

MIT researchers compared brain connectivity patterns among people who wrote essays using ChatGPT, a search engine, or relying solely on their own thinking. The study looks at how reliance on AI tools may influence cognitive engagement during writing tasks. Source: britannica.com