AI News from August 11, 2026

Accel reportedly raised $3.5 billion for emerging global AI startups, signaling continued investor demand despite limited public details on the fund.

NVIDIA is showcasing open-weight models and local AI tools that let developers run agents, coding, video and robotics workloads on its hardware.

NVIDIA launched an efficient open model and routing library aimed at lowering the cost, latency and deployment friction of enterprise AI agents.

H2O.ai has joined NVIDIA’s Open Secure AI Alliance, adding an enterprise AI platform to an effort focused on open tools for safer agents.

Nvidia’s Nemotron 3.5 Lightning uses sparse activation and quantization to deliver fast open-weight inference, targeting high-volume AI agents over peak scores.

NVIDIA's NeMo Switchyard routes AI agent tasks across models to balance accuracy, latency, infrastructure limits, and inference cost in production workflows.

DeepSeek and Unitree are reported to be collaborating on robotics AI models, a move that could link model development with robot hardware.

Nvidia is reportedly working with Wall Street firms to raise $500 billion for AI infrastructure, a plan that could reshape data-center financing.

Meta’s open-weight Glimmer model brings local AI agents to consumer hardware, revealing how the company may split personal AI from its strongest systems.

Meta's new open-weight model is pitched as a way to run AI agents on a laptop, but available reporting leaves its hardware and performance limits unclear.

Two conflicting forecasts say domestic chips could dominate China’s AI market by 2026, raising questions about export controls, supply and demand.