NVIDIA GB300 Grace Blackwell Ultra Superchip 252GB
Performance Metrics
Peak theoretical throughput by precision type
| Precision | Peak TFLOPS | |
|---|---|---|
| FP6 | 5000.0 | |
| FP8 | 5000.0 | |
| FP4 | 10000.0 | |
| FP16 | 2500.0 | |
| BF16 | 2500.0 | |
| TF32 | 1250.0 | |
| FP32 | 80.0 | |
| FP64 | 1.3 |
Power Specifications
TDP
--
Max Power
--
Power Connector
PCIe Slot
Cooling
Air
Memory Specifications
Capacity
252 GB
Type
HBM3e
Bandwidth
7100 GB/s
Interface
--
Hardware & Design
Form Factor
Superchip
Architecture
Blackwell Ultra
Process Node
4nm
Launch Year
2025
Variant
Grace Blackwell Ultra Superchip 252GB
Market Segment
Professional
Full Specifications
| Memory | |
|---|---|
| VRAM | 252 GB |
| Memory Type | HBM3e |
| Bandwidth | 7.1 TB/s |
| Interconnect & I/O | |
| GPU-to-GPU | NVLink-C2C |
| Interconnect Bandwidth | 900 GB/s |
| Power & Thermal | |
| Physical & Media | |
| NVENC Engines | 1 |
| General | |
| Form Factor | Superchip |
| Architecture | Blackwell Ultra |
| Launch Year | 2025 |
Common Use Cases
The NVIDIA GB300 is optimized for high-performance computing tasks with Blackwell Ultra architecture delivering 80 TFLOPS of compute power.
Where to Rent
Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.
Browse GPU Cloud ProvidersSimilar GPUs
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
No spam, unsubscribe anytime.