Releases·Tool Release
NVIDIA Launches cuObject and SCADA Server SDK to Speed AI Access to File and Object Storage
NVIDIA announced cuObject and the NVIDIA SCADA Server SDK, expanding high-speed access to file and object storage for AI workloads. The tools target AI infrastructure engineers, storage developers, and cloud service providers needing fast, secure data access for training, fine-tuning, and inference.
0 engagement1 sourceWed, Sep 30, 2026, 07:13 PM
AI Overview
- NVIDIA targets AI infrastructure engineers, storage developers, and cloud service providers needing fast, secure access to on-premises and cloud file/object storage
- cuObject and the SCADA Server SDK address data access for training, fine-tuning, inference context, tool calls, searches, and database lookups
Related
Releases·Hardware Release
NVIDIA Vera Storage Benchmarks: Faster Encryption, Compression, Integrity Checking, and Recovery for AI-Native Storage
NVIDIA published storage benchmarks for its Vera platform, showing faster encryption, compression, integrity checking, and recovery for AI-native storage. The benchmarks highlight storage's role in agentic AI workflows, where agents retrieve knowledge, access memory, and reuse KV cache data.
NVIDIA published storage benchmarks for its Vera platform, showing faster encryption, compression, integrity checking, and recovery for AI-native storage. The benchmarks highlight storage's role in agentic AI workflows, where agents retrieve knowledge, access memory, and reuse KV cache data.
0 engagement1 sourcerssMon, Aug 3, 2026, 04:00 PM
News·Company News
Enabling Private High-Performance Production AI Inference with NVIDIA Confidential Computing
NVIDIA published a technical post on running production LLM inference inside confidential computing environments, targeting personal, enterprise, and regulated settings where inference handles sensitive data or proprietary model context. The approach uses memory-encrypted confidential virtual machines (CVMs) and confidential GPUs to process workloads in a trusted environment. It gives practitioners a pathway to deploy inference on sensitive data without exposing plaintext memory.
NVIDIA published a technical post on running production LLM inference inside confidential computing environments, targeting personal, enterprise, and regulated settings where inference handles sensitive data or proprietary model context. The approach uses memory-encrypted confidential virtual machines (CVMs) and confidential GPUs to process workloads in a trusted environment. It gives practitioners a pathway to deploy inference on sensitive data without exposing plaintext memory.
0 engagement1 sourcerssTue, Sep 22, 2026, 05:27 PM
News·Company News
How OpenAI scaled Habitat storage to 1 billion ChatGPT users at 22M requests per second
OpenAI published an engineering account of evolving Habitat from a Python library into a globally distributed online storage platform backing ChatGPT. The system now serves over 1 billion users at 22 million requests per second, offering a look at the infrastructure behind large-scale assistant deployments.
OpenAI published an engineering account of evolving Habitat from a Python library into a globally distributed online storage platform backing ChatGPT. The system now serves over 1 billion users at 22 million requests per second, offering a look at the infrastructure behind large-scale assistant deployments.
0 engagement1 sourcerssFri, Sep 11, 2026, 10:00 AM
Releases·Tool Release
NVIDIA releases CUDA MCP server for AI-assisted GPU code and docs
NVIDIA released an NVIDIA-hosted CUDA MCP server that lets AI assistants search official CUDA documentation, write optimized GPU code, and analyze performance data. The tool aims to streamline CUDA development workflows for developers using AI coding assistants.
NVIDIA released an NVIDIA-hosted CUDA MCP server that lets AI assistants search official CUDA documentation, write optimized GPU code, and analyze performance data. The tool aims to streamline CUDA development workflows for developers using AI coding assistants.
83 engagement1 sourceredditThu, Aug 20, 2026, 07:31 PM
Releases·Tool Release
SkyPilot and Hugging Face launch zero-egress storage integration for multi-cloud AI workloads
SkyPilot and Hugging Face announced a joint integration that allows users to mount Hugging Face Hub repos (including Hugging Face Buckets) directly into SkyPilot jobs using an hf:// URL and HF_TOKEN. This enables running AI workloads—development, training, or serving—on any cloud GPU cluster while reading data from the Hub with zero egress fees. The move simplifies multi-cloud AI workflows by decoupling storage from compute.
SkyPilot and Hugging Face announced a joint integration that allows users to mount Hugging Face Hub repos (including Hugging Face Buckets) directly into SkyPilot jobs using an hf:// URL and HF_TOKEN. This enables running AI workloads—development, training, or serving—on any cloud GPU cluster while reading data from the Hub with zero egress fees. The move simplifies multi-cloud AI workflows by decoupling storage from compute.
0 engagement1 sourcerssTue, Jul 7, 2026, 12:00 AM