Grace Blackwell
A high-performance AI accelerator chip introduced by NVIDIA, integrating the Grace CPU with the Blackwell GPU architecture. It is designed for large-scale AI training and inference workloads and is widely deployed in data centers.
Core Facts
Mistral Large 4 was trained using 3,800 NVIDIA Grace Blackwell GPUs in its own data centers, and the public preview also runs on the same infrastructure.
Amazon is seeking to transfer approximately $8 billion worth of Nvidia chip assets to external investors.
Timeline (last 90 days)
At GTC, Nvidia announced that Grace Blackwell products have officially entered full production.
The assembly time of the new generation of Grace Blackwell has been shortened from two hours in the previous generation to five minutes.
The new generation of Grace Blackwell achieves 100% liquid cooling, whereas only 80% of the previous generation used liquid cooling.
Grace Blackwell is NVIDIA's new-generation GPU platform for large-scale AI training, with Grace responsible for the CPU side and Blackwell for the GPU side, forming a superchip through a high-bandwidth interconnect.
ML4's training was completed in Europe, using about 3,800 NVIDIA Grace Blackwell GPUs.
Le Chonk's training used 3,800 NVIDIA Grace Blackwell GPUs located in European data centers.
Mistral Large was trained in European data centers using 3,800 NVIDIA Grace Blackwell chips, with model weights and license to be published by the end of the month.
Mistral Large 4 was trained using 3,800 NVIDIA Grace Blackwell GPUs in its own data centers, and the public preview also runs on the same infrastructure.
Grace Blackwell is Nvidia's latest-generation chip platform for AI training and inference, with significant improvements in energy efficiency and interconnect bandwidth compared to the previous-generation H100
Amazon is seeking to transfer approximately $8 billion worth of Nvidia chip assets to external investors.
5 more timeline events
All Facts (16)
Mistral Large 4 was trained using 3,800 NVIDIA Grace Blackwell GPUs in its own data centers, and the public preview also runs on the same infrastructure.
70%VerifiedAmazon is seeking to transfer approximately $8 billion worth of Nvidia chip assets to external investors.
70%UnverifiedDGX Spark基于NVIDIA的Grace Blackwell架构，采用Grace CPU与Blackwell GPU的紧密耦合设计
90%UnverifiedDGX Spark is a compact AI workstation launched by NVIDIA for individual developers and small teams, featuring the Grace Blackwell chip.
60%UnverifiedThe assembly time of the new generation of Grace Blackwell has been shortened from two hours in the previous generation to five minutes.
50%UnverifiedThe new generation of Grace Blackwell achieves 100% liquid cooling, whereas only 80% of the previous generation used liquid cooling.
50%UnverifiedGrace Blackwell is NVIDIA's new-generation GPU platform for large-scale AI training, with Grace responsible for the CPU side and Blackwell for the GPU side, forming a superchip through a high-bandwidth interconnect.
50%UnverifiedML4's training was completed in Europe, using about 3,800 NVIDIA Grace Blackwell GPUs.
50%UnverifiedLe Chonk's training used 3,800 NVIDIA Grace Blackwell GPUs located in European data centers.
50%UnverifiedMistral Large was trained in European data centers using 3,800 NVIDIA Grace Blackwell chips, with model weights and license to be published by the end of the month.
50%UnverifiedGrace Blackwell is Nvidia's latest-generation chip platform for AI training and inference, with significant improvements in energy efficiency and interconnect bandwidth compared to the previous-generation H100
50%UnverifiedThe GB300 is the latest iteration of NVIDIA's Grace Blackwell architecture, with the previous generation being the GB200.
50%UnverifiedThe GB300 superchip is based on NVIDIA's Grace Blackwell architecture, integrating the CPU and GPU via high-bandwidth interconnects to form a unified memory addressing space.
50%UnverifiedThe Grace Blackwell platform is currently supported primarily through CUDA 12.x and the corresponding cuDNN and NCCL versions, with PyTorch and JAX having optimized kernels for this architecture.
50%UnverifiedThe DGX Spark's Grace Blackwell architecture uses a unified memory pool shared between CPU and GPU, so model weights do not need to be copied between system memory and VRAM, but two large models cannot reside in memory simultaneously.
50%UnverifiedAt GTC, Nvidia announced that Grace Blackwell products have officially entered full production.
50%