OpenAI Says Its AI Agents Solved the Navier-Stokes Millennium Prize Problem
OpenAI released a proof and Lean formalization while framing the result as evidence of faster AI research progress.
OpenAI released a proof and Lean formalization while framing the result as evidence of faster AI research progress.
GPT-6 Astra hit 95% on Robocurve's robot control task against Claude Fable 5.1's 40%, using 6.2x fewer output tokens at 2.3x lower cost.
The OpenAI chief scientist argues that increasingly capable AI systems are becoming harder to understand, monitor, and safely scale.
Nvidia’s founder and CEO says OpenAI’s newest model marks the arrival of artificial general intelligence, while another 400,000 GPUs are expected to come online as the company scales its Stargate infrastructure.
The NVIDIA-backed setup lets Hermes users run open models on their own hardware without manually configuring llama.cpp.
Dyson's first oral-care device uses a camera, machine-learning gap detection, and a mouthrinse jet to target spaces between teeth while users brush.
The Pantheon Release adds multi-agent group chats, persistent cron jobs and a broader MCP management surface to Hermes Agent.
OpenClaw’s largest release to date rebuilds onboarding and its browser app while adding distributed agent sessions, deeper memory, interactive dashboards, automation permissions and new collaboration tools.
The deal gives Nvidia control of one of AI’s most important open-model platforms, extending its reach from the compute layer into model discovery, deployment and an increasingly important robotics ecosystem.
Pocket says it crossed $100 million in annualized run rate seven months after reaching $1 million, as its AI notetaking device scales globally.
Bot Mode turns Hermes Desktop into a multi-agent workspace where users can create persistent specialist bots with their own memory, tools, routines, and chat histories.
Anthropic is rolling out machine-readable watermarks across Claude, including human-written content that is edited or processed by its AI models.
Hermes Agent users can now connect Actual Computer as an inference provider, extending the Bittensor subnet’s private compute model into Nous Research’s open-source agent stack.
The upgraded API dramatically improves coding and agent benchmarks without changing the underlying model, underscoring how inference efficiency is becoming a new competitive battleground.
OpenAI says the lower pricing gives developers and ChatGPT Work users more usage while adding a faster API option for its flagship GPT-5.6 Sol model.
The Bittensor genomics project’s GPU-native synthetic genome engine was highlighted as one of the most ambitious case studies in a new report on AI-assisted scientific software.