LayerStoRm, an open-source engine that runs large MoE LLMs on consumer GPUs using PCIe streaming, has been released.
FreeToken proposes a system for efficiently serving frontier-grade MoE models on hardware comparable to gaming PCs.
The self-improving agent 'Prime Agent', which applies RLM and Continual Harness for coding and long-horizon autonomous tasks, has been released.
The 'Scaffold CoT' dataset, comprising 4 million entries, has been released to improve the reasoning accuracy of small models with 5B parameters or fewer.
Memory is enabled by default for Free, Pro, and Max plans, while sensitive topics require explicit user activation.
The partnership builds on Bain's internal rollout of Claude to 19,000 employees, where over 7,000 adopted it within weeks.
Anthropic has hired Amir Salek, who led Google's TPU division, to build its own silicon organization.
Goodfire has launched a grant program offering up to $1 million in free Silico access to AI interpretability researchers.
GitLab analyzes that as AI makes code production cheap, reliability and governance become the core focus.
AI hardware bottlenecks are cascading from GPUs to data centers, causing build costs to rise exponentially.
Rome runs persistent AI agents and workflows in a guardrailed collaborative environment.
It has been raised that LLMs can control host machines by exploiting code execution vulnerabilities in inference engines.
Nvidia has disclosed the hardware requirements for running CUDA on RISC-V server CPUs.
The value of AI models varies according to the intelligence required for each task, with excessive intelligence leading to inefficiency.
The anonymous AI model Ox Alpha broke launch records by processing 26 trillion tokens on OpenCode in just 4 days.
NVIDIA has begun mass production of the Groq 3 LPX, significantly boosting inference speeds on the Vera Rubin platform.
MiniMax M3 and M2.7 models are available for free on Vercel AI Gateway until September 6.
Alibaba's Wan 3.0 video model has launched on Vercel AI Gateway, offering integrated video generation capabilities.
OpenAI has released an Admin plugin for managing ChatGPT Work and Codex workspaces.
Vercel Connect is now generally available, enabling secure connections to over 100 services using short-lived tokens at runtime.
Vercel AI Gateway has added asynchronous video generation capabilities via webhooks and polling.
Apple released STARFlow2, which combines LLMs and normalizing flows to generate text and images in a unified manner.
The Chat SDK's Slack adapter now supports Slack Enterprise Grid, enabling integration across all workspaces in an organization.
Vercel Connect added support for the Linq connector, enabling iMessage and SMS-based messaging capabilities.
AI Briefing