Thousands of NVIDIA Grace Blackwell GPUs Now Live at CoreWeave Boosting AI Development
CoreWeave deploys thousands of NVIDIA Blackwell GPUs in GB200 NVL72 systems for AI pioneers like Cohere IBM and Mistral AI to train and run advanced AI models
CoreWeave has become one of the first cloud providers to deploy NVIDIA GB200 NVL72 systems at scale, with AI leaders like Cohere, IBM, and Mistral AI already leveraging these powerful GPUs for next-generation AI model training and deployment.
Thousands of NVIDIA Blackwell GPUs are now operational at CoreWeave, with more coming online soon.
Key Developments:
-
CoreWeave's Milestone: The cloud provider is the first to make NVIDIA Grace Blackwell generally available, showcasing impressive MLPerf benchmarks with the NVIDIA GB200 NVL72 platform.
-
AI Factory Transformation: These systems are turning cloud data centers into AI factories, enabling real-time insights with unprecedented speed and efficiency.
Company-Specific Implementations:
Cohere: Personalized AI Agents
Cohere is using Grace Blackwell Superchips to develop its North platform, which enables enterprises to build secure, personalized AI agents. Early results show:
- 3x performance boost in training 100B-parameter models compared to previous-gen Hopper GPUs
- Dramatically higher throughput for inference
IBM: Enterprise AI Models
IBM is scaling to thousands of Blackwell GPUs on CoreWeave to train its Granite models, with key benefits:
- State-of-the-art performance for enterprise AI applications
- Integration with IBM watsonx Orchestrate for workflow automation
- Combined with IBM Storage Scale System for high-performance AI storage
Mistral AI: Open-Source Innovation
Paris-based Mistral AI has received its first thousand Blackwell GPUs, achieving:
- 2x performance improvement in dense model training out-of-the-box
- Enhanced capabilities for developing open-source language models
Future Outlook
CoreWeave now offers instances with:
- Rack-scale NVIDIA NVLink across 72 GPUs and 36 CPUs
- Scalability to 110,000 GPUs with NVIDIA Quantum-2 InfiniBand
This infrastructure is set to accelerate the development of AI reasoning models and agentic AI applications across industries.
Related News
Lenovo Wins Frost Sullivan 2025 Asia-Pacific AI Services Leadership Award
Lenovo earns Frost Sullivan's 2025 Asia-Pacific AI Services Customer Value Leadership Recognition for its value-driven innovation and real-world AI impact.
Baidu Wenku GenFlow 2.0 Revolutionizes AI Agents with Multi-Agent Architecture
Baidu Wenku's GenFlow 2.0 introduces a multi-agent system for parallel task processing, integrating with Cangzhou OS to enhance efficiency and redefine AI workflows.
About the Author
Alex Thompson
AI Technology Editor
Senior technology editor specializing in AI and machine learning content creation for 8 years. Former technical editor at AI Magazine, now provides technical documentation and content strategy services for multiple AI companies. Excels at transforming complex AI technical concepts into accessible content.