A powerful new open-source framework called vla-evaluation-harness launches with the ability to evaluate 500+ robot AI models across 17 benchmarks at up to 47x faster speeds, completing 2,000 simulation episodes in just 18 minutes on a single GPU.
Nvidia CEO Jensen Huang launches NemoClaw, a free open-source AI agent platform, repositioning Nvidia as the operating system of the AI era while using the move to defend GPU hardware demand against rival chips and prevent customers like OpenAI from gaining enough dominance to threaten Nvidia's pricing power.
Open source maintainers are drowning in AI-generated pull requests, and a new '3 Cs' framework — Comprehension, Context, and Continuity — is emerging as an urgent solution to help them filter genuine contributors, reduce burnout, and protect the mentorship culture that keeps open source communities thriving.
Microsoft launches an open-source Agent Package Manager (APM) that lets developers declare all AI agent dependencies in a single apm.yml file, with built-in security auditing, transitive dependency resolution, and support for GitHub, GitLab, and Azure DevOps across macOS, Linux, and Windows.
NVIDIA launches the Nemotron Coalition, uniting Mistral AI, Perplexity, LangChain, and other leading AI labs to co-develop open, frontier-level AI foundation models, with the first project trained on NVIDIA DGX Cloud set to power the upcoming Nemotron 4 model family.
NousResearch launches Hermes Agent, an open-source, self-improving AI that learns from experience, supports 200+ LLMs, runs on infrastructure as cheap as a $5 VPS, and connects across Telegram, Discord, WhatsApp, Signal, and Slack — already earning 8.7k GitHub stars from 142 contributors.
AWS launches llm-d disaggregated inference on SageMaker HyperPod and EKS, delivering up to 70% higher LLM throughput by splitting compute-heavy prefill and memory-bound decode phases across distributed GPUs using an open-source Kubernetes-native framework built on vLLM.
Google Labs launches Stitch SDK, an open-source TypeScript library that lets developers instantly generate UI screens from simple text prompts, complete with HTML extraction, screenshot support, and seamless AI framework integration.
Physical Superintelligence PBC launches 'Get Physics Done,' the first open-source agentic AI physicist, featuring 61 commands, a four-stage research workflow, and support for four major AI runtimes, now available on GitHub and installable with a single command.
Andrej Karpathy releases an open-source tool on GitHub that scores 342 US occupations on AI exposure risk using an interactive treemap and an LLM-powered pipeline, empowering anyone to fork and customize the analysis for insights beyond just digital disruption.