All posts

nvidia

8 posts

NVIDIA Labs Introduces SoL-Pi for Coding Agent Efficiency
AgentsSep 11, 2026·4 min read

NVIDIA Labs Introduces SoL-Pi for Coding Agent Efficiency

NVIDIA Labs released SoL-Pi, an open-source extension for the Pi coding agent harness that reduces token usage and operational costs via recursive research.

Read more →
NVIDIA Introduces CUDA Rust with SIMT and Tile Programming Tracks
ToolsSep 9, 2026·4 min read

NVIDIA Introduces CUDA Rust with SIMT and Tile Programming Tracks

NVIDIA announced native CUDA programming in Rust via two distinct tracks: the SIMT-based cuda-oxide and the Tile-based cutile-rs compiler toolchains.

Read more →
NVIDIA Launches NemoClaw for LangChain Deep Agents Code: Run Open-source dcode, tuned for Nemotron 3 Ultra, to Plan, Edit, and Test Code
AgentsJul 18, 2026·5 min read

NVIDIA Launches NemoClaw for LangChain Deep Agents Code: Run Open-source dcode, tuned for Nemotron 3 Ultra, to Plan, Edit, and Test Code

NVIDIA NemoClaw provides a governed blueprint for deploying LangChain’s Deep Agents Code in secure, sandboxed environments for sensitive software engineering.

Read more →
NVIDIA Launches Verified Agent Skills to Supercharge AI Coders
AgentsJun 24, 2026·4 min read

NVIDIA Launches Verified Agent Skills to Supercharge AI Coders

NVIDIA has released an open-source catalog of verified Agent Skills, enabling AI assistants like Cursor and Claude to seamlessly utilize NVIDIA SDKs without hallucinating.

Read more →
Google Releases DiffusionGemma: 4x Faster Text Generation via Diffusion
Open SourceJun 10, 2026·4 min read

Google Releases DiffusionGemma: 4x Faster Text Generation via Diffusion

Google's experimental DiffusionGemma model ditches autoregressive generation for text diffusion, achieving up to 4x faster local inference speeds on GPUs.

Read more →
NVIDIA Unveils Nemotron 3 Ultra 550B MoE for AI Agents
AnnouncementsJun 6, 2026·4 min read

NVIDIA Unveils Nemotron 3 Ultra 550B MoE for AI Agents

NVIDIA's new 550B-parameter Nemotron 3 Ultra MoE model brings hybrid Mamba-Transformer layers and MOPD training to long-running agent orchestration.

Read more →
NVIDIA Cosmos 3: The World's First Fully Open Omnimodel for Physical AI
AI ModelsJun 1, 2026·5 min read

NVIDIA Cosmos 3: The World's First Fully Open Omnimodel for Physical AI

NVIDIA has announced Cosmos 3 at GTC Taipei COMPUTEX — the world's first fully open omnimodel for Physical AI, combining native vision reasoning, world generation, and action simulation in a single model. Available now in Super (32B) and Nano (8B) variants.

Read more →
NVIDIA Launches Verified Agent Skills: Transparency and Security for Agentic AI Pipelines
AI ToolsMay 22, 2026·4 min read

NVIDIA Launches Verified Agent Skills: Transparency and Security for Agentic AI Pipelines

NVIDIA has released Verified Agent Skills — a system for adding transparency, provenance, and risk metadata to AI agent skills. Each verified skill carries a skill card detailing what it does, where it came from, and whether it's been modified, built on the open agentskills.io specification and compatible with Claude Code, OpenAI Codex, and Cursor.

Read more →